CN104484372A - Detecting method and device of business object sending information - Google Patents

Abstract

The invention provides a method and a device for detecting information delivered by a business object. Wherein, the method includes: generating a detection model of business object delivery information based on the attribute information of a plurality of historically uploaded service object delivery information deliverers in advance; receiving a detection request carrying a delivery provider identifier; according to the detection model of business object delivery information, The delivery information of the business object uploaded by the deliverer corresponding to the deliverer identifier is detected. The detection model of the business object delivery information of the present invention is generated in combination with a large amount of historical information analysis, so it can more objectively and accurately detect the business object delivery information uploaded by the publisher, and there is no need to check the detected information when performing detection according to the detection model. The delivery effect of the advertiser is analyzed and counted, the detection process is more convenient, and the detection efficiency is improved.

Description

A method and device for detecting information delivered by a business object

technical field

The invention relates to the field of network technology, in particular to a method and device for detecting information delivered by a business object.

Background technique

With the continuous development of the Internet, there are more and more network users, so more and more publishers put business objects on the Internet business object delivery platform, so as to recommend business objects to these network users, so that users can understand in real time to the latest information.

The advertiser participates in the competition by purchasing keywords and bidding to cover as much traffic as possible and increase the number of clicks, formulate business object placement information according to business structure, region, time, etc., and upload the business object placement information to the business object placement platform, the business object placement platform performs business object placement according to the business object placement information.

The pros and cons of the information placed by the business object will have a direct impact on the coverage traffic and click volume of the advertiser. If the quality of the information delivered by the business object is poor, it will lead to the poor effect of the advertiser’s advertising on the business object and affect the income. Especially for some inexperienced advertisers, due to their lack of experience in formulating business object placement information and keyword bidding, etc., it often leads to the situation that the bidding ranking is low or the bidding ranking environment is disturbed by excessive bidding. Therefore, if it is possible to detect the delivery information of the business object uploaded by the publisher, it is of great significance to remind the publisher of problems existing in the delivery information of the business object and to formulate more optimized delivery information of the business object.

The current method of detecting the delivery information of business objects uploaded by the publisher is mainly to analyze the historical delivery effect of the delivery information of the business objects uploaded by the publisher separately, and generate corresponding analysis reports, such as analyzing different regions and different interests , The historical delivery effect of different keywords, so as to generate regional reports, interest reports, keyword reports, etc.

However, using the above-mentioned business object delivery information detection method that separately analyzes the historical delivery effect of the business object delivery information uploaded by the publisher, the detection result is inaccurate, the detection efficiency is low, and it cannot meet the needs of users.

Contents of the invention

In view of the above problems, the present invention is proposed to provide a method for detecting business object placement information and a corresponding detection device for business object placement information that overcome the above problems or at least partially solve the above problems.

According to one aspect of the present invention, a method for detecting service object delivery information is provided, including:

Generate a detection model for business object delivery information in advance based on the attribute information of multiple publishers who have previously uploaded business object delivery information;

Receive a detection request carrying the identifier of the advertiser;

According to the detection model of the delivery information of the business object, the delivery information of the delivery service object uploaded by the delivery provider corresponding to the delivery provider identifier is detected.

According to another aspect of the present invention, a device for detecting service object delivery information is provided, including:

The generation module is adapted to generate a detection model of business object delivery information based on the attribute information of the publishers who have previously uploaded business object delivery information;

The receiving module is adapted to receive the detection request carrying the identifier of the advertiser;

The detection module is adapted to detect the delivery information of the service object uploaded by the deliverer corresponding to the deliverer identifier according to the detection model of the delivery information of the service object.

According to the detection scheme of the business object delivery information of the present invention, a detection model of the business object delivery information is generated based on the attribute information of a plurality of deliverers who have previously uploaded the service object delivery information in advance, and then after receiving the detection request carrying the identifier of the deliverer After that, the service object delivery information uploaded by the provider corresponding to the provider identifier can be detected according to the generated detection model of the service object delivery information. The above-mentioned detection model of business object delivery information is generated by combining a large amount of historical information analysis, so it can more objectively and accurately detect the business object delivery information uploaded by the provider, and it is not necessary to check the detected provider when performing detection based on the detection model. The delivery effect is analyzed and counted, the detection process is more convenient, and the detection efficiency is improved.

The above description is only an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable , the specific embodiments of the present invention are enumerated below.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiment. The drawings are only for the purpose of illustrating a preferred embodiment and are not to be considered as limiting the invention. Also throughout the drawings, the same reference numerals are used to designate the same components. In the attached picture:

FIG. 1 shows a flow chart of the steps of a method for detecting information delivered by a business object in Embodiment 1 of the present invention;

FIG. 2 shows a flow chart of steps of a method for detecting information delivered by a business object in Embodiment 2 of the present invention;

Fig. 3 shows a structural block diagram of a device for detecting service object delivery information in Embodiment 3 of the present invention;

Fig. 4 shows a structural block diagram of a device for detecting service object delivery information in Embodiment 4 of the present invention.

Detailed ways

Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided for more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

Embodiment one:

Referring to FIG. 1 , it shows a flow chart of steps of a method for detecting service object placement information in Embodiment 1 of the present invention. The method may include the steps of:

In step 100, a detection model of business object delivery information is generated in advance based on the attribute information of a plurality of deliverers who have previously uploaded business object delivery information.

In the embodiment of the present invention, firstly, the attribute information of the provider who uploaded the service object delivery information based on multiple historical uploads can be obtained, and then the detection model of the service object delivery information can be generated based on these attribute information, and then the detection of the delivery information of the business object can be performed subsequently The model detects the delivery information of the business object uploaded by the deliverer.

Step 102, receiving a detection request carrying the identifier of the advertiser.

When a certain publisher requests to detect the delivery information of the business object uploaded by it, corresponding operations may be performed, and these operations may trigger the generation of a detection request, and the detection request may include the identifier of the publisher requesting detection.

Step 104 , according to the detection model of the service object delivery information, detect the service object delivery information uploaded by the delivery provider corresponding to the delivery provider identifier.

After receiving the above-mentioned detection request carrying the provider ID, the service object delivery information uploaded by the provider corresponding to the provider ID can be detected according to the detection model of the pre-generated service object delivery information, so as to obtain the service The scoring value of object delivery information provides a basis for locating problems existing in business object delivery information and formulating more optimized business object delivery information.

In the embodiment of the present invention, the detection model of business object delivery information is generated in combination with a large amount of historical information analysis, so it can be more objective and accurate to detect the business object delivery information uploaded by the publisher, and it is not necessary to perform detection based on the detection model. Analyzing and counting the delivery effects of the detected advertisers makes the detection process more convenient and improves the detection efficiency.

Embodiment two:

Referring to FIG. 2 , it shows a flow chart of steps of a method for detecting service object placement information in Embodiment 2 of the present invention. The method may include the steps of:

In step 200, a detection model of business object delivery information is generated in advance based on the attribute information of a plurality of deliverers who have previously uploaded business object delivery information.

When advertisers need to recommend business objects to network users, they first formulate a promotion plan, divide promotion groups, purchase keywords, set creative ideas, etc., then generate business object placement information based on these factors, and finally log in to the business object delivery platform and place the business object The delivery information is uploaded to the business object delivery platform. The subsequent business object delivery platform can launch the business object according to the business object delivery information.

In the embodiment of the present invention, the detection model of business object delivery information can be generated based on the attribute information of multiple publishers who have uploaded business object delivery information in history. Therefore, the detection model can be more objectively Detect the delivery information of the business object.

In a preferred embodiment of the present invention, this step 200 may include the following sub-step a1 to sub-step a3:

Sub-step a1, pre-collect multiple display logs and multiple click logs in the browser;

When a user browses a webpage in a browser, after entering a keyword to search, the search engine will refer to information such as bids for the keyword by different business objects to sort the business objects corresponding to the keyword, and decide whether to push the keyword to the left, and then Display these business objects in search results pages. When a web page is displayed, a display log will be generated in the browser, which describes the behavior of displaying the web page. The display log may include: the identifier of the displayed business object, the identifier of the publisher to which the displayed business object belongs, information about whether the displayed business object has been pushed to the left, and the quality score of the displayed business object (calculated by the search engine) Information related to the displayed business objects, etc. Of course, the displayed log may also include other displayed information, such as related information about displayed keywords, related information about displayed pictures, and so on.

When a user clicks a business object displayed on a web page, a click log will be generated in the browser, and the click log describes the behavior of clicking the business object. The click log may include: the identifier of the clicked business object, the identifier of the publisher to which the clicked business object belongs, information about whether the clicked business object has been pushed to the left, the consumption value of the clicked business object, etc. Information about the clicked business object.

Sub-step a2, according to the display log and the click log, respectively count the attribute information of the puter of each historically uploaded business object delivery information;

Among them, the attribute information of the publisher may include: pageviews, and/or left pageviews, and/or clicks, and/or left clicks, and/or quality scores, and/or consumption values, and/or The click rate, that is, the attribute information of the publisher may include any one or more of page views, left page views, clicks, left clicks, quality score, consumption value, and click rate.

Therefore, this sub-step a2 may include the following sub-steps:

Step a21, counting the number of displayed business objects belonging to the same publisher in all display logs, and using this number as the number of views of the publisher; and/or,

Step a22, counting the number of display business objects that belong to the same publisher and are pushed to the left in all display logs, and use this number as the left view of the publisher; and/or,

Step a23, counting the number of clicked business objects belonging to the same publisher in all click logs, and using this number as the click volume of the publisher; and/or,

Step a24, counting the number of business objects that belong to the same publisher and are pushed to the left in all click logs, and use this number as the volume of left clicks of the publisher; and/or,

Step a25, counting the number of business objects displayed on the same display in all display logs, and using this number as the number of views of the business object; counting the number of business objects clicked on the same in all click logs, and using this number as the number of views of the business object The number of hits of the object; respectively calculate the quotient of the hits and views of each business object, and use the quotient as the click-through rate of the business object; calculate the average of the click-through rates of all business objects belonging to the same publisher value as the CTR for said trafficker; and/or,

Step a26, calculating the average value of quality scores of displayed business objects belonging to the same provider in all display logs, and using the average value as the quality score of the provider; and/or,

Step a27, calculating the sum of the consumption values of the clicked business objects belonging to the same publisher in all the click logs, and taking the sum as the consumption value of the publisher.

According to the content included in the attribute information of the advertiser, the corresponding sub-step is selected from the above-mentioned sub-step a21 to sub-step a27 to execute. For example, when the publisher's attribute information includes page views, left page views, clicks, and left clicks, step a21 to substep a24 are performed, and so on. When all or part of the sub-steps in the above step a21 to sub-step a27 are executed, there is no limitation on the execution order of each sub-step, which can be executed successively or simultaneously.

After counting the attribute information of the publishers of the delivery information of each historically uploaded business object, the attribute information of these publishers can also be written into the database for use in subsequent queries.

In sub-step a3, a detection model of business object delivery information is generated based on the attribute information of the multiple deliverers.

After the attribute information of each of the above-mentioned publishers is acquired, a detection model of business object delivery information can be generated based on the attribute information of the above-mentioned multiple publishers.

In a preferred embodiment of the present invention, the sub-step a3 may include the following sub-steps a31-sub-step a34:

Step a31, for each attribute information of each provider, respectively calculate the ranking score value of the current attribute information in all the attribute information of the provider that is the same as the attribute information;

The attribute information of multiple providers is counted, and each provider includes one or more attribute information. Therefore, for each attribute information of each provider, the current attribute information is calculated separately for all the providers with the same attribute information. Rank score value in attribute information. For example, for the views of one of the publishers, calculate the ranking score value of the traffic of this trafficker among the traffic of all traffickers; The rank score value in the click volume of the user. After the calculation, the ranking score value of each attribute information of each provider can be obtained.

Wherein, the ranking score value may be a standardized percentile system, and the ranking score value may be calculated according to the ratio of the ranking of the current attribute information to the total number of contributors. For example, if the rank of the current attribute information is 100 and the total number of publishers is 1000, then the ranking of the current attribute information accounts for 10% of the total number of publishers, and the ranking score value of the current attribute information is 90 points.

Step a32, using the ranking score values of all attribute information of a provider as an object, clustering all objects;

Taking the ranking score values of all attribute information of a provider as an object, each object may correspond to a feature vector, and the feature vector is a feature vector composed of the ranking score values of all attribute information of the provider.

The sub-step a32 may include the following sub-steps a321-sub-step a326:

Sub-step a321, perform hierarchical clustering on all objects, and determine the initial clustering of the target number;

This sub-step a321 may include the following sub-steps a3211-sub-step a3215:

Sub-step a3211, using an object as an initial cluster, calculate the distance between every two initial clusters;

Calculating the distance between two initial clusters is to calculate the distance between the feature vectors corresponding to the two initial clusters.

If an object is an initial cluster, the distance between every two initial clusters is the distance between the feature vectors corresponding to the two objects. In the embodiment of the present invention, calculating the distance between two feature vectors may be calculating the Euclidean distance, Manhattan distance, cosine similarity, Hamming distance, Minnesota distance, etc. between the two feature vectors. Euclidean distance comes from the distance formula between two points in Euclidean space, the distance between two n-dimensional vectors a(x ₁₁ , x ₁₂ ,…, x _1n ) and b(x ₂₁ , x ₂₂ ,…, x _2n ) The Euclidean distance is: It can also be expressed as a vector operation: Manhattan distance is also called city block distance. The Manhattan distance between two n-dimensional vectors a(x ₁₁ , x ₁₂ ,…,x _1n ) and b(x ₂₁ , x ₂₂ ,…,x _2n ) is: For the calculation process of the remaining distances, this embodiment of the present invention will not discuss one by one here.

Sub-step a3212, merging the two initial clusters with the smallest distance into one initial cluster;

Sub-step a3213, using the following formula to calculate the B(k) value corresponding to the initial clustering:

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; interDis</span>}<span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; intraDis</span>}$

Among them, interDis is the distance between each two initial clusters, intraDis is the sum of the distances between each two objects within the initial cluster, and k is the number of initial clusters;

Through this sub-step a3213, the B(k) value corresponding to each initial cluster can be calculated.

Sub-step a3214, calculate the distance between the merged initial cluster and each other initial cluster, and return to the step of merging the two initial clusters with the smallest distance into one initial cluster, until the initial cluster until the number is 1;

If an initial cluster includes at least two objects, when calculating the distance between the initial cluster and other initial clusters, the corresponding feature vector of the initial cluster is the center point of the initial cluster, that is, the Average of the eigenvectors corresponding to all objects included in the initial cluster.

After the sub-step a3214, a corresponding B(k) value can be obtained for the current initial clustering situation after each combination.

Sub-step a3215, find the minimum B(k) value among all B(k) values, and determine the k initial clusters corresponding to the minimum B(k) values as the target number of initial clusters.

Sub-step a322, randomly select the centroid of each initial cluster;

For each initial cluster, the centroid of the initial cluster is randomly selected from the objects included in the initial cluster, and the centroid is the feature vector corresponding to the randomly selected object.

Sub-step a323, for each object, calculate the distance between the current object and each centroid, and classify the current object into the cluster corresponding to the centroid with the smallest distance between the objects;

The distance between the current object and the centroid is the distance between the feature vector corresponding to the current object and the centroid. For example, if the initial clustering of the determined number of targets is 6, then there are 6 centroids correspondingly, the distances between the current object and the above 6 centroids are calculated respectively, and the current object is classified to the centroid with the smallest distance to the object in the corresponding cluster.

Sub-step a324, judging whether the obtained cluster corresponding to each centroid satisfies the convergence condition; if not, execute sub-step a325; if yes, execute sub-step a326;

This sub-step a324 may include the following sub-steps a3241-sub-step a3243:

Sub-step a3241, using the following formula to calculate the obtained A value corresponding to the cluster corresponding to each centroid:

$\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; min</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; I</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \&Element;</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; dist</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; center</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, I is the number of clusters, C _i is the combination of objects in the i-th cluster, x _j is the j-th object in the i-th cluster, center(i) is the center of the i-th cluster, The center of the i-th cluster is the average value of all objects in the i-th cluster;

dist(center(i), x _j ) is the distance between center(i) of the i-th cluster and the j-th object x _j in the i-th cluster, that is, all The distance between the mean of the eigenvectors corresponding to an object and the eigenvector corresponding to the jth object in the ith cluster.

Sub-step a3242, obtain the A value calculated by the previous preset times, and compare the A value calculated this time with every two adjacent A values in the A value calculated by the previous preset times;

In sub-step a3243, if the variation ranges of every two adjacent A values are within the preset range, then it is determined that the obtained cluster corresponding to each centroid satisfies the convergence condition.

Wherein, the embodiment of the present invention does not limit the specific values of the preset times and the preset range.

Sub-step a325, recalculate the centroid of each initial cluster, and return to sub-step a323;

If it is judged in sub-step a324 that the convergence condition is not met, then recalculate the centroid of each initial cluster. Here, recalculating the centroid of the initial cluster may be calculating an average value of all objects included in the initial cluster, that is, an average value of feature vectors corresponding to all included objects. Then return to execute sub-step a323.

Sub-step a326, determine the obtained cluster corresponding to each centroid as the clustering result.

Sub-step a33, for each cluster, perform linear regression analysis respectively to obtain the regression parameters corresponding to the current cluster;

After the above sub-step a32, the clusters corresponding to each centroid are obtained, and then linear regression analysis is performed on each cluster to obtain the regression parameters corresponding to the current cluster.

This sub-step a33 may include the following sub-steps a331-sub-step a332:

Sub-step a331, for each cluster, respectively determine the following formula corresponding to each object in the current cluster:

Y _n ＝β ₀ +β ₁ X ₁ +β ₂ X ₂ +…+β _m X _m +e

Among them, β ₀ ~ β _m are regression parameters, X ₁ ~ X _m are the ranking score values of the attribute information of the nth object respectively, Y _n is the ranking score value of the consumption value of the nth object, m is the nth object The quantity of attribute information of the object, e is a random error;

In sub-step a332, the values of the regression parameters β ₀ -β _m corresponding to the current cluster are calculated according to the equation system composed of formulas corresponding to each object in the current cluster.

Each object in the current cluster corresponds to one of the above formulas. For the current cluster, the values of β ₀ ~ β _m can be calculated according to the equations composed of these methods. During the calculation process, the random error e is can be deleted.

According to the sub-step a33, the regression parameters corresponding to each cluster can be determined.

In the embodiment of the present invention, the sub-step a33 can use SPSS (Statistical Product and Service Solutions) software to perform linear regression analysis on the clustering results. Certainly, other manners may also be used, which are not limited in this embodiment of the present invention.

Step a34, for each cluster, respectively use the regression parameters corresponding to the current cluster to determine the detection model of the service object delivery information corresponding to the cluster.

This sub-step a34 may include the following sub-steps a341-sub-step a342:

In sub-step a341, for each cluster, use the values of the regression parameters β ₀ ~ β _m corresponding to the current cluster to calculate the following formula:

$\mathrm{<span\; class="notranslate">\; Score</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}$

In sub-step a342, the calculated formula is determined as a detection model for the delivery information of the business object corresponding to the cluster.

Therefore, a detection model of the delivery information of the business object corresponding to each cluster can be obtained.

In a preferred embodiment of the present invention, after the above sub-step a33 is performed, for each cluster, a significance test and/or a back-substitution test can be further performed on the regression parameters corresponding to the current cluster. That is, only the significance test can be carried out, or only the back-substitution test can be carried out, or both the significance test and the back-substitution test can be carried out.

The significance test is to make a hypothesis about the parameters or the overall distribution form of the population (random variable) in advance, and then use the sample information to judge whether the hypothesis (alternate hypothesis) is reasonable, that is, to judge whether the real situation of the population is significantly different from the original hypothesis. sexual difference. In the above sub-step a33, for each cluster, after linear regression analysis is performed separately, the coefficient of determination and the coefficient of determination can be automatically obtained. Whether the coefficients are greater than the preset threshold; if yes, it is determined to pass the significance test; if not, it is determined to fail the significance test.

The process of back-substitution inspection is: count the attribute information of at least one provider who uploaded the delivery information of the business object (it can be counted according to the above steps a1 to sub-step a2), and for each attribute information of each provider, respectively Calculate the ranking score value of the current attribute information in the attribute information identical to the attribute information of all contributors; Substitute the ranking score value of the attribute information of at least one provider into the formula in the above-mentioned sub-step a331 respectively (at this time, β ₀ ～ β _m is known, and e is deleted), calculates the corresponding Y value (the ranking score value of the consumption value); calculates the variance of the ranking score value of the above-mentioned Y value and the actual consumption value of the object; judges whether the variance Within the preset range; if yes, it is determined to pass the back-substitution test; if not, it is determined to fail the back-substitution test.

If the regression parameters corresponding to a certain cluster do not pass the significance test or the back-substitution test, you can delete the cluster, or return to perform sub-step a32; if the significance test and the back-substitution test pass, then execute this sub-step a4.

Step 202, receiving a detection request carrying the identifier of the advertiser.

In the embodiment of the present invention, a detection control can be pre-set in the browser. When a publisher needs to detect the service object delivery information uploaded by himself, he can trigger and generate a detection request by clicking the detection control. The detection request may include the trafficker's ID for that trafficker.

Step 204 , according to the detection model of the service object delivery information, detect the service object delivery information uploaded by the delivery provider corresponding to the delivery provider identifier.

After receiving the detection request carrying the identifier of the publisher, the delivery information of the business object uploaded by the publisher corresponding to the identifier of the publisher is detected according to the detection model of the delivery information of the business object.

In a preferred embodiment of the present invention, this step 204 may include the following sub-step b1 to sub-step b4:

Sub-step b1, obtaining the attribute information of the puter corresponding to the puter identifier;

First, multiple display logs and multiple click logs in the browser may be collected; then, according to the display logs and click logs, the attribute information of the provider corresponding to the provider identifier is counted. For the specific process, refer to the relevant descriptions of the above-mentioned sub-step a1 to sub-step a2.

Sub-step b2, determining the cluster to which the attribute information of the advertiser corresponding to the identifier belongs;

This step b2 may include the following sub-steps b21 to b23:

Sub-step b21, for each attribute information of the advertiser corresponding to the identifier, respectively calculate the ranking score value of the current attribute information;

In this sub-step b21, firstly, based on the display log and click log obtained in the above-mentioned sub-step b1, the attribute information of each publisher can be counted respectively; Calculate the ranking score value of the current attribute information. For the specific process, refer to the relevant descriptions of the above-mentioned sub-step a2 to sub-step a3.

Sub-step b22, using the ranking score value of the attribute information of the advertiser as an object, respectively calculating the distance between the object and the centroid corresponding to each cluster;

Sub-step b23, determining that the cluster corresponding to the centroid with the smallest distance between the objects is the cluster to which the attribute information of the puter corresponding to the identifier belongs.

Sub-step b3, using the ranking score value of the attribute information of the advertiser corresponding to the identifier as the input of the detection model of the business object delivery information corresponding to the determined cluster;

After determining the cluster to which the attribute information of the provider corresponding to the identifier belongs, the detection model of the delivery information of the business object corresponding to the drama category can be obtained, which is the following formula:

$\mathrm{<span\; class="notranslate">\; Score</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}$

Then, the ranking score value of the attribute information identifying the corresponding provider is used as the input of the detection model of the service object's delivery information, that is, X ₁ -X _m .

In sub-step b4, the output of the detection model of the delivery information of the business object is used as the score value of the delivery information of the business object uploaded by the delivery provider corresponding to the delivery identifier.

After the calculation of the detection model of the delivery information of the business object, the final output of the detection model of the delivery information of the business object is the scoring value of the delivery information of the business object uploaded by the delivery provider corresponding to the delivery identifier.

Step 206, display the detection result of the service object placement information uploaded by the puter corresponding to the puter identifier.

After the above detection result is obtained, the detection result of the delivery information of the business object uploaded by the delivery provider corresponding to the delivery identifier can be displayed in the browser, that is, the score value of the delivery information of the uploaded business object.

Step 208, acquire attribute information of the advertiser corresponding to the identifier of the advertiser, and display the attribute information.

In a preferred embodiment of the present invention, the attribute information of the advertiser corresponding to the identifier of the advertiser can be further acquired, and the attribute information can be displayed.

The embodiment of the present invention does not limit the specific display manner, for example, it may be displayed by loading a prompt box, and the prompt box includes score values and the like.

By displaying the score value and attribute information of the delivery information of the business object corresponding to the delivery provider to the delivery provider, the delivery provider can more clearly understand the delivery effect of its business object, and to locate the problems existing in the delivery information of the business object and formulate more optimized Provide the basis for the delivery information of the business object.

Step 206 and step 208 are not limited to the above-mentioned execution sequence, step 206 can be executed first, and then step 208 can be executed; step 208 can also be executed first, and then step 206 can be executed; step 206 and step 208 can also be executed at the same time, the embodiment of the present invention This is not limited.

The business object in the embodiment of the present invention may be an advertisement, the business object delivery platform is the advertisement delivery platform, the placer is the advertiser, and the service object delivery information is the advertisement delivery strategy.

In the embodiment of the present invention, firstly, an objective health score is made based on indicators such as the attribute information of the advertisers, which provides the main basis for subsequent formulation of optimization suggestions; secondly, combined with the delivery effects of a large number of advertisers, the overall bidding environment is improved. degree, which is beneficial to the overall revenue of the business object delivery platform; finally, using multiple linear regression analysis to build a detection model can more accurately describe the relationship between the health score of the launcher and each indicator.

Embodiment three:

Referring to FIG. 3 , it shows a structural block diagram of a device for detecting service object delivery information in Embodiment 3 of the present invention.

The generation module 300 is adapted to generate a detection model of business object delivery information based on the attribute information of a plurality of historically uploaded service object delivery information deliverers in advance;

The receiving module 302 is adapted to receive the detection request carrying the identifier of the advertiser;

The detecting module 304 is adapted to detect the delivery information of the service object uploaded by the deliverer corresponding to the deliverer identifier according to the detection model of the delivery information of the service object.

In the embodiment of the present invention, the detection model of the service object delivery information is generated based on the attribute information of the deliverer who has previously uploaded the service object delivery information in advance, and after receiving the detection request carrying the delivery person's identifier, the generated The detection model of the delivery information of the business object detects the delivery information of the business object uploaded by the delivery provider corresponding to the delivery identifier. The above-mentioned detection model of business object delivery information is generated by combining a large amount of historical information analysis, so it can more objectively and accurately detect the business object delivery information uploaded by the provider, and it is not necessary to check the detected provider when performing detection based on the detection model. The delivery effect is analyzed and counted, the detection process is more convenient, and the detection efficiency is improved.

Embodiment four:

Referring to FIG. 4 , it shows a structural block diagram of a device for detecting service object placement information in Embodiment 4 of the present invention.

The generation module 400 is adapted to generate a detection model of business object delivery information based on the attribute information of a plurality of historically uploaded service object delivery information deliverers in advance;

The receiving module 402 is adapted to receive the detection request carrying the identifier of the advertiser;

The detecting module 404 is adapted to detect the delivery information of the service object uploaded by the deliverer corresponding to the deliverer identifier according to the detection model of the delivery information of the service object.

The result display module 406 is adapted to display the detection result of the service object placement information uploaded by the puter corresponding to the puter identifier;

The attribute displaying module 408 is adapted to acquire attribute information of the puter corresponding to the puter identifier, and display the attribute information.

In a preferred embodiment of the present invention, the generating module may include the following submodules:

The log collection sub-module is suitable for pre-collecting multiple display logs and multiple click logs in the browser;

The attribute statistics sub-module is adapted to count the attribute information of each historical uploader of the service object delivery information according to the display log and the click log;

The model generation sub-module is adapted to generate a detection model of business object delivery information based on the attribute information of the multiple deliverers.

Wherein, the display log includes: the identifier of the displayed business object, the identifier of the advertiser to which the displayed business object belongs, information about whether the displayed business object is pushed to the left, and the quality score of the displayed business object The click log includes: the identifier of the clicked business object, the identifier of the publisher to which the clicked business object belongs, information about whether the clicked business object is pushed to the left, and the consumption value of the clicked business object; The attribute information of the publisher includes: views, and/or left views, and/or clicks, and/or left clicks, and/or quality score, and/or consumption value, and/or clicks Rate.

The attribute statistics submodule can include the following units:

The pageview counting unit is adapted to count the number of displayed business objects belonging to the same publisher in all display logs, and use this number as the pageview volume of the publisher; and/or,

The left pageview statistics unit is suitable for counting the number of displayed business objects that belong to the same publisher and are pushed to the left in all display logs, and use this number as the left pageview of the publisher; and/or,

The click statistics unit is suitable for counting the number of clicked business objects belonging to the same publisher in all click logs, and using this number as the click volume of the publisher; and/or,

The left click volume statistics unit is suitable for counting the number of business objects that belong to the same publisher and are pushed to the left in all click logs, and use this number as the left click volume of the publisher; and/or,

The click-through rate statistics unit is suitable for counting the number of the same displayed business objects in all display logs, and using this number as the pageview volume of the business object; counting the number of the same clicked business objects in all click logs, and using the number as the click volume of the business object; respectively calculate the quotient value of the click volume and page view volume of each business object, and use the quotient value as the click rate of the business object; calculate the quotient of all business objects belonging to the same publisher the average of the click-through rates as the click-through rate for said advertiser; and/or,

The score statistics unit is adapted to calculate the average value of the quality scores of the displayed business objects belonging to the same provider in all the display logs, and use the average value as the quality score of the provider; and/or,

The consumption statistics unit is adapted to calculate the sum of the consumption values of the clicked business objects belonging to the same publisher in all the click logs, and use the sum as the consumption value of the publisher.

The model generation submodule can include the following units:

The calculation unit is adapted to calculate, for each attribute information of each poster, the ranking score value of the current attribute information among all the attribute information of the poster that is the same as the attribute information;

The clustering unit is suitable for clustering all objects by using the ranking score values of all attribute information of a provider as an object;

The analysis unit is adapted to perform linear regression analysis on each cluster to obtain regression parameters corresponding to the current cluster;

The determination unit is adapted to use the regression parameters corresponding to the current cluster to determine the detection model of the service object delivery information corresponding to the cluster for each cluster.

Among them, the clustering unit can include the following sub-units:

Hierarchical clustering subunit, suitable for hierarchical clustering of all objects, to determine the initial clustering of the target number;

select subunits, suitable for randomly selecting the centroid of each initial cluster;

The classification subunit is adapted to calculate the distance between the current object and each centroid for each object, and classify the current object into the cluster corresponding to the centroid with the smallest distance between the objects;

The judging subunit is suitable for judging whether the cluster corresponding to each obtained centroid satisfies the convergence condition; if not, recalculate the centroid of each initial cluster, and call the classification subunit; if so, determine the obtained The cluster corresponding to each centroid is the clustering result.

Hierarchical clustering subunit, specifically for:

Take an object as an initial cluster, and calculate the distance between each two initial clusters;

Merge the two initial clusters with the smallest distance into one initial cluster;

Utilize the following formula to calculate the B(k) value corresponding to the initial clustering:

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; interDis</span>}<span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; intraDis</span>}$

Among them, interDis is the distance between each two initial clusters, intraDis is the sum of the distances between each two objects within the initial cluster, and k is the number of initial clusters;

Calculate the distance between the merged initial cluster and each other initial cluster, and return the step of merging the two initial clusters with the smallest distance into one initial cluster until the number of initial clusters is 1 until;

Find the minimum B(k) value among all B(k) values, and determine the k initial clusters corresponding to the minimum B(k) values as the target number of initial clusters.

The judging subunit is specifically suitable for:

Use the following formula to calculate the A value corresponding to the cluster corresponding to each centroid:

$\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; min</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; I</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \&Element;</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; dist</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; center</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, I is the number of clusters, C _i is the combination of objects in the i-th cluster, x _j is the j-th object in the i-th cluster, center(i) is the center of the i-th cluster, The center of the i-th cluster is the average value of all objects in the i-th cluster;

Get the A value calculated by the previous preset times, and compare the A value calculated this time with every two adjacent A values in the A value calculated by the previous preset times;

If the change range of every two adjacent A values is within the preset range, it is determined that the obtained cluster corresponding to each centroid satisfies the convergence condition.

Analysis units can include the following subunits:

The formula determination subunit is suitable for determining the following formula corresponding to each object in the current cluster for each cluster:

Y _n ＝β ₀ +β ₁ X ₁ +β ₂ X ₂ +…+β _m X _m +e

Among them, β ₀ ~ β _m are regression parameters, X ₁ ~ X _m are the ranking score values of the attribute information of the nth object respectively, Y _n is the ranking score value of the consumption value of the nth object, m is the nth object The quantity of attribute information of the object, e is a random error;

The parameter calculation subunit is adapted to calculate the values of the regression parameters β ₀ ˜β _m corresponding to the current cluster according to an equation system composed of formulas corresponding to each object in the current cluster.

A determination unit may include the following subunits:

The formula calculation subunit is suitable for calculating the following formula for each cluster using the regression parameters β ₀ ~ β _m corresponding to the current cluster:

$\mathrm{<span\; class="notranslate">\; Score</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}$

The clustering model determination subunit is adapted to determine the calculated formula as a detection model for the delivery information of the business object corresponding to the cluster.

The detection module can include the following submodules:

The attribute acquisition sub-module is adapted to acquire the attribute information of the advertiser corresponding to the identifier of the advertiser;

The cluster determination submodule is adapted to determine the cluster to which the attribute information of the advertiser corresponding to the identifier belongs;

The information scoring sub-module is adapted to use the ranking score value of the attribute information of the provider corresponding to the identification as the input of the detection model of the business object delivery information corresponding to the determined cluster; the input of the detection model of the business object delivery information Output as the score value of the service object placement information uploaded by the puter corresponding to the puter identifier.

The cluster determination submodule may include the following units:

The score calculation unit is adapted to calculate the ranking score value of the current attribute information for each attribute information of the advertiser corresponding to the identifier;

The distance calculation unit is adapted to use the ranking score value of the attribute information of the advertiser as an object, and calculate the distance between the object and the centroid corresponding to each cluster;

The cluster determination unit is adapted to determine that the cluster corresponding to the centroid with the smallest distance between the objects is the cluster to which the attribute information of the puter corresponding to the identifier belongs.

The above-mentioned detection model of business object delivery information is generated by combining a large amount of historical information analysis, so it can more objectively and accurately detect the business object delivery information uploaded by the provider, and it is not necessary to check the detected provider when performing detection based on the detection model. The delivery effect is analyzed and counted, the detection process is more convenient, and the detection efficiency is improved.

The algorithms and displays presented herein are not inherently related to any particular computer, virtual system, or other device. Various generic systems can also be used with the teachings based on this. The structure required to construct such a system is apparent from the above description. Furthermore, the present invention is not specific to any particular programming language. It should be understood that various programming languages can be used to implement the contents of the present invention described herein, and the above description of specific languages is for disclosing the best mode of the present invention.

In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure the understanding of this description.

Similarly, it should be appreciated that in the foregoing description of exemplary embodiments of the invention, in order to streamline this disclosure and to facilitate an understanding of one or more of the various inventive aspects, various features of the invention are sometimes grouped together in a single embodiment, figure, or its description. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.

Those skilled in the art can understand that the modules in the device in the embodiment can be adaptively changed and arranged in one or more devices different from the embodiment. Modules or units or components in the embodiments may be combined into one module or unit or component, and furthermore may be divided into a plurality of sub-modules or sub-units or sub-assemblies. All features disclosed in this specification (including accompanying claims, abstract and drawings) and any method or method so disclosed may be used in any combination, except that at least some of such features and/or processes or units are mutually exclusive. All processes or units of equipment are combined. Each feature disclosed in this specification (including accompanying claims, abstract and drawings) may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise.

Furthermore, those skilled in the art will understand that although some embodiments described herein include some features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention. and form different embodiments. For example, in the following claims, any of the claimed embodiments may be used in any combination.

The various component embodiments of the present invention may be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. Those skilled in the art should understand that a microprocessor or a digital signal processor (DSP) may be used in practice to implement some or all functions of some or all of the components in the detection device for placing information on a business object according to an embodiment of the present invention . The present invention can also be implemented as an apparatus or an apparatus program (for example, a computer program and a computer program product) for performing a part or all of the methods described herein. Such a program for realizing the present invention may be stored on a computer-readable medium, or may be in the form of one or more signals. Such a signal may be downloaded from an Internet site, or provided on a carrier signal, or provided in any other form.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The use of the words first, second, and third, etc. does not indicate any order. These words can be interpreted as names.

A1. A method for detecting information delivered by a business object, including:

Generate a detection model for business object delivery information in advance based on the attribute information of multiple publishers who have previously uploaded business object delivery information;

Receive a detection request carrying the identifier of the advertiser;

According to the detection model of the delivery information of the business object, the delivery information of the delivery service object uploaded by the delivery provider corresponding to the delivery provider identifier is detected.

A2. The method as described in A1, wherein the step of generating a detection model of business object delivery information based on the attribute information of a plurality of historically uploaded service object delivery information providers includes:

Pre-collect multiple display logs and multiple click logs in the browser;

According to the display log and the click log, respectively count the attribute information of the sender of each historically uploaded business object delivery information;

A detection model of business object delivery information is generated based on the attribute information of the plurality of deliverers.

A3. The method as described in A2, wherein,

The display log includes: the identifier of the displayed business object, the identifier of the advertiser to which the displayed business object belongs, information about whether the displayed business object is pushed to the left, and the quality score of the displayed business object;

The click log includes: the identifier of the clicked business object, the identifier of the publisher to which the clicked business object belongs, information about whether the clicked business object is pushed to the left, and the consumption value of the clicked business object;

The attribute information of the publisher includes: views, and/or left views, and/or clicks, and/or left clicks, and/or quality score, and/or consumption value, and/or clicks Rate.

A4. The method as described in A3, wherein the step of counting the attribute information of each historically uploaded service object delivery information provider according to the display log and the click log includes:

Count the number of business objects displayed in all impression logs belonging to the same dropper as the number of pageviews for said dropper; and/or,

Count the number of business objects of impressions that belong to the same deliverer and are pushed to the left in all impression logs, and use this number as the left view of the deliverer; and/or,

Count the number of clicked business objects belonging to the same publisher in all click logs, and use this number as the click volume of the publisher; and/or,

Count the number of business objects that belong to the same publisher and are pushed to the left in all click logs, and use this number as the volume of left clicks of the publisher; and/or,

Count the number of the same displayed business object in all display logs, and use this number as the number of views of the business object; count the number of the same clicked business object in all click logs, and use this number as the number of clicks on the business object amount; respectively calculate the quotient value of the hits and views of each business object, and use the quotient as the click-through rate of the business object; calculate the average value of the click-through rates of all business objects belonging to the same publisher as the click-through rate of said advertiser; and/or,

calculating the average of the Quality Scores of the Business Objects in the Impression Log belonging to the same dropper for the impression as the Quality Score for said dropper; and/or,

Calculate the sum of the consumption values of the clicked business objects belonging to the same publisher in all the click logs, and use the sum as the consumption value of the publisher.

A5. The method as described in A2, wherein the step of generating a detection model of business object delivery information based on the attribute information of the multiple deliverers includes:

For each attribute information of each provider, calculate the ranking score value of the current attribute information among all the attribute information of the provider that is the same as the attribute information;

Use the ranking score value of all attribute information of a provider as an object to cluster all objects;

For each cluster, linear regression analysis is performed separately to obtain the regression parameters corresponding to the current cluster;

For each cluster, the regression parameters corresponding to the current cluster are used to determine the detection model of the service object delivery information corresponding to the cluster.

A6. The method as described in A5, wherein the step of clustering all objects comprises:

Hierarchical clustering of all objects, determining an initial cluster of the target number;

Randomly select the centroid of each initial cluster;

For each object, calculate the distance between the current object and each centroid, and classify the current object into the cluster corresponding to the centroid with the smallest distance between the objects;

Judging whether the cluster corresponding to each centroid obtained satisfies the convergence condition;

If not, recalculate the centroid of each initial cluster, and return the above. For each object, calculate the distance between the current object and each centroid, and classify the current object to the object with the smallest distance The centroids correspond to the steps in the clustering;

If so, determine the cluster corresponding to each centroid obtained as the clustering result.

A7. The method as described in A6, wherein the step of performing hierarchical clustering on all objects and determining the initial clustering of the target quantity includes:

Take an object as an initial cluster, and calculate the distance between each two initial clusters;

Merge the two initial clusters with the smallest distance into one initial cluster;

Utilize the following formula to calculate the B(k) value corresponding to the initial clustering:

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; interDis</span>}<span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; intraDis</span>}$

Among them, interDis is the distance between each two initial clusters, intraDis is the sum of the distances between each two objects within the initial cluster, and k is the number of initial clusters;

Calculate the distance between the merged initial cluster and each other initial cluster, and return the step of merging the two initial clusters with the smallest distance into one initial cluster until the number of initial clusters is 1 until;

Find the minimum B(k) value among all B(k) values, and determine the k initial clusters corresponding to the minimum B(k) values as the target number of initial clusters.

A8. The method as described in A6, wherein the step of judging whether the obtained cluster corresponding to each centroid satisfies the convergence condition includes:

Use the following formula to calculate the A value corresponding to the cluster corresponding to each centroid:

$\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; min</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; I</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \&Element;</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; dist</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; center</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, I is the number of clusters, C _i is the combination of objects in the i-th cluster, x _j is the j-th object in the i-th cluster, center(i) is the center of the i-th cluster, The center of the i-th cluster is the average value of all objects in the i-th cluster;

Get the A value calculated by the previous preset times, and compare the A value calculated this time with every two adjacent A values in the A value calculated by the previous preset times;

If the change range of every two adjacent A values is within the preset range, it is determined that the obtained cluster corresponding to each centroid satisfies the convergence condition.

A9. The method according to A5, wherein the attribute information includes a consumption value,

The step of performing linear regression analysis for each cluster to obtain the regression parameters corresponding to the current cluster includes:

For each cluster, determine the following formulas for each object in the current cluster:

Y _n ＝β ₀ +β ₁ X ₁ +β ₂ X ₂ +…+β _m X _m +e

Among them, β ₀ ~ β _m are regression parameters, X ₁ ~ X _m are the ranking score values of the attribute information of the nth object respectively, Y _n is the ranking score value of the consumption value of the nth object, m is the nth object The quantity of attribute information of the object, e is a random error;

Calculate the values of the regression parameters β ₀ ~ β _m corresponding to the current cluster according to the equation system composed of formulas corresponding to each object in the current cluster.

A10. The method as described in A9, wherein, for each cluster, the steps of using the regression parameters corresponding to the current cluster to determine the detection model of the service object delivery information corresponding to the cluster include:

For each cluster, use the values of the regression parameters β ₀ ~ β _m corresponding to the current cluster to calculate the following formula:

$\mathrm{<span\; class="notranslate">\; Score</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}$

The calculated formula is determined as the detection model of the delivery information of the business object corresponding to the cluster.

A11. The method as described in A6, wherein, according to the detection model of the delivery information of the business object, the step of detecting the delivery information of the delivery service object uploaded by the delivery provider corresponding to the delivery provider identification includes:

Acquiring the attribute information of the provider corresponding to the provider identifier;

determining the cluster to which the attribute information of the advertiser corresponding to the identifier belongs;

Using the ranking score value of the attribute information of the advertiser corresponding to the identifier as the input of the detection model of the business object delivery information corresponding to the determined cluster;

The output of the detection model of the delivery information of the business object is used as the scoring value of the delivery information of the business object uploaded by the delivery provider corresponding to the delivery identifier.

A12. The method according to A11, wherein the step of determining the cluster to which the attribute information of the advertiser corresponding to the identifier belongs includes:

For each attribute information of the advertiser corresponding to the identifier, calculate the ranking score value of the current attribute information;

Taking the ranking score value of the attribute information of the publisher as an object, calculating the distance between the object and the centroid corresponding to each cluster;

The cluster corresponding to the centroid with the smallest distance between the objects is determined as the cluster to which the attribute information of the puter corresponding to the identifier belongs.

A13. The method as described in A1, further comprising:

The detection result of the delivery information of the business object uploaded by the deliveryr corresponding to the deliveryr identifier is displayed.

A14. The method as described in A1, further comprising:

Obtain the attribute information of the provider corresponding to the provider identifier, and display the attribute information.

B15. A detection device for information delivery by a business object, including:

The generation module is adapted to generate a detection model of business object delivery information based on the attribute information of the publishers who have previously uploaded business object delivery information;

The receiving module is adapted to receive the detection request carrying the identifier of the advertiser;

The detection module is adapted to detect the service object delivery information uploaded by the provider corresponding to the provider identifier according to the detection model of the service object delivery information.

B16. The device as described in B15, wherein the generating module includes:

The log collection sub-module is suitable for pre-collecting multiple display logs and multiple click logs in the browser;

The attribute statistics sub-module is adapted to count the attribute information of each historical uploader of the service object delivery information according to the display log and the click log;

The model generation sub-module is adapted to generate a detection model of business object delivery information based on the attribute information of the multiple deliverers.

B17. The device of B16, wherein,

The display log includes: the identifier of the displayed business object, the identifier of the advertiser to which the displayed business object belongs, information about whether the displayed business object is pushed to the left, and the quality score of the displayed business object;

The click log includes: the identifier of the clicked business object, the identifier of the publisher to which the clicked business object belongs, information about whether the clicked business object is pushed to the left, and the consumption value of the clicked business object;

The attribute information of the publisher includes: views, and/or left views, and/or clicks, and/or left clicks, and/or quality score, and/or consumption value, and/or clicks Rate.

B18. The device as described in B17, wherein the attribute statistics submodule includes:

The pageview counting unit is adapted to count the number of displayed business objects belonging to the same publisher in all display logs, and use this number as the pageview volume of the publisher; and/or,

The left pageview statistics unit is suitable for counting the number of displayed business objects that belong to the same publisher and are pushed to the left in all display logs, and use this number as the left pageview of the publisher; and/or,

The click statistics unit is suitable for counting the number of clicked business objects belonging to the same publisher in all click logs, and using this number as the click volume of the publisher; and/or,

The left click volume statistics unit is suitable for counting the number of business objects that belong to the same publisher and are pushed to the left in all click logs, and use this number as the left click volume of the publisher; and/or,

The click-through rate statistics unit is suitable for counting the number of the same displayed business objects in all display logs, and using this number as the pageview volume of the business object; counting the number of the same clicked business objects in all click logs, and using the number as the click volume of the business object; respectively calculate the quotient value of the click volume and page view volume of each business object, and use the quotient value as the click rate of the business object; calculate the quotient of all business objects belonging to the same publisher the average of the click-through rates as the click-through rate for said advertiser; and/or,

The score statistics unit is adapted to calculate the average value of the quality scores of the displayed business objects belonging to the same provider in all display logs, and use the average value as the quality score of the provider; and/or,

The consumption statistics unit is adapted to calculate the sum of the consumption values of the clicked business objects belonging to the same publisher in all the click logs, and use the sum as the consumption value of the publisher.

B19. The device as described in B16, wherein the model generation submodule includes:

The calculation unit is adapted to calculate, for each attribute information of each poster, the ranking score value of the current attribute information among all the attribute information of the poster that is the same as the attribute information;

The clustering unit is suitable for clustering all objects by using the ranking score values of all attribute information of a provider as an object;

The analysis unit is adapted to perform linear regression analysis on each cluster to obtain regression parameters corresponding to the current cluster;

The determination unit is adapted to use the regression parameters corresponding to the current cluster to determine the detection model of the service object delivery information corresponding to the cluster for each cluster.

B20, the device as described in B19, wherein, the clustering unit comprises:

Hierarchical clustering subunit, suitable for hierarchical clustering of all objects, to determine the initial clustering of the target number;

select subunits, suitable for randomly selecting the centroid of each initial cluster;

The classification subunit is adapted to calculate the distance between the current object and each centroid for each object, and classify the current object into the cluster corresponding to the centroid with the smallest distance between the objects;

The judging subunit is suitable for judging whether the cluster corresponding to each obtained centroid satisfies the convergence condition; if not, recalculate the centroid of each initial cluster, and call the classification subunit; if so, determine the obtained The cluster corresponding to each centroid is the clustering result.

B21. The device as described in B20, wherein the hierarchical clustering subunit is specifically adapted to:

Take an object as an initial cluster, and calculate the distance between each two initial clusters;

Merge the two initial clusters with the smallest distance into one initial cluster;

Utilize the following formula to calculate the B(k) value corresponding to the initial clustering:

$\mathrm{<span\; class="notranslate">\; B</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; interDis</span>}<span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}\mathrm{<span\; class="notranslate">\; intraDis</span>}$

Among them, interDis is the distance between each two initial clusters, intraDis is the sum of the distances between each two objects within the initial cluster, and k is the number of initial clusters;

Calculate the distance between the merged initial cluster and each other initial cluster, and return the step of merging the two initial clusters with the smallest distance into one initial cluster until the number of initial clusters is 1 until;

Find the minimum B(k) value among all B(k) values, and determine the k initial clusters corresponding to the minimum B(k) values as the target number of initial clusters.

B22. The device as described in B20, wherein the judging subunit is specifically adapted to:

Use the following formula to calculate the A value corresponding to the cluster corresponding to each centroid:

$\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; min</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; I</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}_{{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; \&Element;</span>{\mathrm{<span\; class="notranslate">\; C</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}}\mathrm{<span\; class="notranslate">\; dist</span>}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; center</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}<span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

Among them, I is the number of clusters, C _i is the combination of objects in the i-th cluster, x _j is the j-th object in the i-th cluster, center(i) is the center of the i-th cluster, The center of the i-th cluster is the average value of all objects in the i-th cluster;

Get the A value calculated by the previous preset times, and compare the A value calculated this time with every two adjacent A values in the A value calculated by the previous preset times;

If the change range of every two adjacent A values is within the preset range, it is determined that the obtained cluster corresponding to each centroid satisfies the convergence condition.

B23. The device as described in B19, wherein the attribute information includes a consumption value, and the analysis unit includes:

The formula determination subunit is suitable for determining the following formula corresponding to each object in the current cluster for each cluster:

Y _n ＝β ₀ +β ₁ X ₁ +β ₂ X ₂ +…+β _m X _m +e

Among them, β ₀ ~ β _m are regression parameters, X ₁ ~ X _m are the ranking score values of the attribute information of the nth object respectively, Y _n is the ranking score value of the consumption value of the nth object, m is the nth object The quantity of attribute information of the object, e is a random error;

The parameter calculation subunit is adapted to calculate the values of the regression parameters β ₀ ˜β _m corresponding to the current cluster according to an equation system composed of formulas corresponding to each object in the current cluster.

B24. The device as described in B23, wherein the determining unit includes:

The formula calculation subunit is suitable for calculating the following formula for each cluster using the regression parameters β ₀ ~ β _m corresponding to the current cluster:

$\mathrm{<span\; class="notranslate">\; Score</span>}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}{{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&beta;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}}$

The clustering model determination subunit is adapted to determine the calculated formula as a detection model for the delivery information of the business object corresponding to the cluster.

B25. The device as described in B20, wherein the detection module includes:

The attribute acquisition sub-module is adapted to acquire the attribute information of the advertiser corresponding to the identifier of the advertiser;

The cluster determination submodule is adapted to determine the cluster to which the attribute information of the advertiser corresponding to the identifier belongs;

The information scoring sub-module is adapted to use the ranking score value of the attribute information of the provider corresponding to the identification as the input of the detection model of the business object delivery information corresponding to the determined cluster; the input of the detection model of the business object delivery information Output as the score value of the service object placement information uploaded by the puter corresponding to the puter identifier.

B26. The device as described in B25, wherein the cluster determination submodule includes:

The score calculation unit is adapted to calculate the ranking score value of the current attribute information for each attribute information of the advertiser corresponding to the identifier;

The distance calculation unit is adapted to use the ranking score value of the attribute information of the advertiser as an object, and calculate the distance between the object and the centroid corresponding to each cluster;

The cluster determination unit is adapted to determine that the cluster corresponding to the centroid with the smallest distance between the objects is the cluster to which the attribute information of the puter corresponding to the identifier belongs.

B27. The device as described in B15, further comprising:

The result display module is adapted to display the detection result of the service object delivery information uploaded by the provider corresponding to the provider identifier.

B28. The device as described in B15, further comprising:

The attribute display module is adapted to obtain the attribute information of the puter corresponding to the puter identifier, and display the attribute information.

Claims (10)

1. a detection method for business object impression information, is characterized in that, comprising:

Upload the attribute information of the putting person of business object impression information in advance based on multiple history, generate the detection model of business object impression information;

Receive the detection request carrying putting person's mark;

According to the detection model of described business object impression information, the business object impression information that corresponding putting person uploads is identified to described putting person and detects.

2. the method for claim 1, is characterized in that, the described attribute information uploading the putting person of business object impression information in advance based on multiple history, and the step generating the detection model of business object impression information comprises:

Many of gathering in advance in browser show daily record and many click logs;

The attribute information that each history uploads the putting person of business object impression information is added up respectively according to described displaying daily record and described click logs;

Attribute information based on described multiple putting person generates the detection model of business object impression information.

3. method as claimed in claim 2, is characterized in that,

Described displaying daily record comprises: whether the business object of the mark of the putting person that the mark of the business object of displaying, the business object of described displaying belong to, described displaying is pushed away the massfraction of the information on a left side, the business object of described displaying;

Described click logs comprises: whether the business object of the mark of the putting person that the mark of the business object of click, the business object of described click belong to, described click is pushed away the consumption figures of the information on a left side, the business object of described click;

The attribute information of described putting person comprises: pageview, and/or left side pageview, and/or click volume, and/or left side click volume, and/or massfraction, and/or consumption figures, and/or clicking rate.

4. method as claimed in claim 3, is characterized in that, describedly adds up according to described displaying daily record and described click logs the step that each history uploads the attribute information of the putting person of business object impression information respectively and comprises:

Add up in all displaying daily records the quantity of business object that belong to same putting person, that show, using the pageview of this quantity as described putting person; And/or,

Add up in all displaying daily records and belong to same putting person and pushed away the quantity of the business object of left displaying, using the left side pageview of this quantity as described putting person; And/or,

Add up in all click logs the quantity of business object that belong to same putting person, that click, using the click volume of this quantity as described putting person; And/or,

Add up in all click logs and belong to same putting person and pushed away the quantity of the business object of left click, using the left side click volume of this quantity as described putting person; And/or,

Add up the quantity of the business object of same displaying in all displaying daily records, using the pageview of this quantity as described business object; Add up the quantity of the business object of same click in all click logs, using the click volume of this quantity as described business object; Calculate the click volume of each business object and the quotient of pageview respectively, using the clicking rate of described quotient as described business object; Calculate the mean value belonging to the clicking rate of all business objects of same putting person, as the clicking rate of described putting person; And/or,

Calculate in all displaying daily records the mean value of the massfraction of business object that belong to same putting person, that show, using the massfraction of this mean value as described putting person; And/or,

Calculate in all click logs the summation of the consumption figures of business object that belong to same putting person, that click, using the consumption figures of this summation as described putting person.

5. method as claimed in claim 2, is characterized in that, the step that the described attribute information based on described multiple putting person generates the detection model of business object impression information comprises:

For each attribute information of each putting person, calculate the Rank scores value of current attribute information in the attribute information that all putting persons are identical with this attribute information respectively;

Using the Rank scores value of all properties information of a putting person as an object, cluster is carried out to all objects;

For each cluster, carry out linear regression analysis respectively, obtain the regression parameter that current cluster is corresponding;

For each cluster, regression parameter corresponding to current cluster is utilized to determine the detection model of the business object impression information that this cluster is corresponding respectively.

6. method as claimed in claim 5, is characterized in that, describedly comprises the step that all objects carry out cluster:

Hierarchical clustering is carried out to all objects, determines the initial clustering of destination number;

The barycenter of each initial clustering of random selecting;

For each object, calculate the distance between existing object and each barycenter respectively, and existing object is referred in cluster corresponding to the barycenter minimum with the spacing of this object;

Whether the cluster that each barycenter that judgement obtains is corresponding meets the condition of convergence;

If not, then recalculate the barycenter of each initial clustering, and return described for each object, calculate the distance between existing object and each barycenter respectively, and existing object is referred to the step in cluster corresponding to the barycenter minimum with the spacing of this object;

If so, then determine that the cluster that each barycenter of obtaining is corresponding is cluster result.

7. method as claimed in claim 6, is characterized in that, describedly carries out hierarchical clustering to all objects, determines that the step of the initial clustering of destination number comprises:

Using an object as an initial clustering, calculate the distance between every two initial clusterings respectively;

An initial clustering is merged into by apart from minimum two initial clusterings;

Utilize B (k) value that initial clustering described in following formulae discovery is corresponding:

$B\left(k\right)=\underset{1}{\overset{C_k^2}{\mathrm{\&Sigma;}}}\mathrm{interDis}+\underset{1}{\overset{k}{\mathrm{\&Sigma;}}}\mathrm{intraDis}$

Wherein, interDis is the distance between every two initial clusterings, and intraDis is the distance sum between inner every two objects of initial clustering, and k is the quantity of initial clustering;

Calculating the distance between the initial clustering after merging and other each initial clusterings, and return and describedly will merge into the step of an initial clustering apart from minimum two initial clusterings, is till 1 until the number of initial clustering;

Search minimum B (k) value in all B (k) values, k corresponding for described minimum B (k) value initial clustering is defined as the initial clustering of destination number.

8. method as claimed in claim 6, it is characterized in that, the step whether cluster that each barycenter that described judgement obtains is corresponding meets the condition of convergence comprises:

Utilize the A value corresponding to cluster that each barycenter of obtaining described in following formulae discovery is corresponding:

$A=\min {\mathrm{\&Sigma;}}_{i=1}^I{\mathrm{\&Sigma;}}_{x_j\&Element;C_i}\mathrm{dist}{(\mathrm{center}\left(i\right),x_j)}^2$

Wherein, I is the quantity of cluster, C _ibe the combination of object in i-th cluster, x _jbe the jth object in i-th cluster, center (i) is the center of i-th cluster, and the center of i-th cluster is the mean value of all objects in i-th cluster;

The A value that before obtaining, preset times calculates, and two adjacent A values every in the A value of this calculating and the A value that calculates of front preset times are compared;

If the amplitude of variation of every two adjacent A values is all in preset range, then determine that the cluster that each barycenter of obtaining is corresponding meets the condition of convergence.

9. method as claimed in claim 5, it is characterized in that, described attribute information comprises consumption figures,

Described for each cluster, carry out linear regression analysis respectively, the step obtaining regression parameter corresponding to current cluster comprises:

For each cluster, determine the following formula that each object in current cluster is corresponding respectively:

Y _n＝β ₀+β ₁X ₁+β ₂X ₂+…+β _mX _m+e

Wherein, β ₀~ β _mfor regression parameter, X ₁~ X _mbe respectively the Rank scores value of the attribute information of the n-th object, Y _nbe the Rank scores value of the consumption figures of the n-th object, m is the quantity of the attribute information of the n-th object, and e is stochastic error;

The system of equations that the formula corresponding according to each object in current cluster forms calculates regression parameter β corresponding to current cluster ₀~ β _mvalue.

10. a pick-up unit for business object impression information, is characterized in that, comprising:

Generation module, is suitable for the attribute information of the putting person uploading business object impression information in advance based on multiple history, generates the detection model of business object impression information;

Receiver module, is suitable for receiving the detection request carrying putting person's mark;

Detection module, is suitable for the detection model according to described business object impression information, identifies the business object impression information that corresponding putting person uploads detect described putting person.