CN103365904B - A kind of advertising message searching method and system - Google Patents

Abstract

The application provides an advertisement information search method and system, which relate to the field of network technology. The method of the present application includes: expanding the query words input by the client to obtain an expanded set of candidate words; analyzing the real-time log of the search engine and/or the advertiser status information at the current time to obtain a valid word set; the advertiser status information Including the keyword display condition information selected by the advertiser; the effective word set includes keywords that meet the display conditions at the current time; using the effective word set, filter out the words in the candidate word set that are not in the effective word set; select At least one candidate word in the filtered candidate word set retrieves advertisement information. This application filters out the extended candidate words that cannot recall advertisements in real time, ensuring that the search keywords input to the search engine at the current time can recall advertisements for display in time, without affecting the display efficiency of advertisements, and can make full use of limited display resources for display advertise.

Description

Method and system for searching advertisement information

technical field

The present application relates to the field of network technology, in particular to a method and system for searching advertisement information.

Background technique

With the development of the Internet, the information provided by the Internet has become more and more diversified, and advertising on the Internet has become an important choice for advertisers. The service of advertising on the Internet can not only promote advertisers, but also provide users with More server information. As an information server platform, Internet search engines have become an advertising delivery method generally adopted by advertisers through search engines.

When Internet users search, they submit query words to the search engine, and the search engine returns the links of a group of webpages most relevant to the query words to the client as search result pages, and users can click on these links to visit related webpages. In the prior art, when placing an advertisement through a search engine, the advertiser registers a group of search keywords closely related to the advertisement link website or web page to be promoted through the search keyword bidding ranking system; First, the query words are expanded to obtain candidate words, and then the candidate words most relevant to the query words are selected as search keywords among the candidate words, and then searched according to each expanded search keyword, if the keyword registered by the advertiser is the same as the search word If the above search keywords match, then it can be displayed. For example, enter the query word "Nokia mobile phone", which can expand "Nokia", "mobile phone", "Nokia N97", "Nokia 5230" and other candidate words, and then select from these candidate words the correlation with "Nokia mobile phone" is greater than the threshold Words are used as search keywords to search.

But in the prior art of the actual existing technology: firstly, there is an upper limit to the number of search words that a search engine can receive, if the number of search words exceeds this upper limit, the efficiency and performance of the search engine will be greatly affected, That is, when the search engine expands based on the query words, the number of expanded search keywords it selects will not exceed the maximum number it can receive; second, for the location of the advertisement, in practice, it can effectively display The placement of advertisements is limited; thirdly, for each advertiser, in practice, advertisers usually limit the display of a set of search keywords selected by themselves (such as the display time period) according to the actual situation, then, if the search The engine searches for the advertiser's advertisement in the advertisement library according to the expanded search keywords, but at this time it does not meet the display conditions set by the client (for example, it is not within the display time period set by the client), then the search engine will The advertisement will not be recalled and placed in the advertising information search position for delivery.

Based on the foregoing, the prior art has the following disadvantages:

In the prior art, the retrieval keywords expanded by the search engine based on the query words input by the customer are only based on correlation considerations, the optimal number of search words that can be received by the search engine, the number of effective advertisement information search positions, and the number of each advertisement information. Under the customer's restrictions on the display conditions of its own advertisements, there may be one or several search keywords in the extended search keywords, and the advertisements cannot be recalled for display after searching the keyword advertisement index library, and because the search engine can receive There is a limit on the number of search keywords. If the search engine expands the search keywords that cannot recall advertisements, the words will occupy the search resources, causing other search words related to the query words to be unable to enter the search process in time or not, affecting the search results. The display of other advertisements that are related to the query words and meet the display conditions prevents the expanded search keywords from being able to fully utilize the limited advertisement positions to display the advertiser's advertisements.

Contents of the invention

The technical problem to be solved by this application is to provide an advertisement information search method and system, which solves the problems in the prior art that the expanded words cannot be effectively recalled for display after searching the keyword advertisement index database, and the display efficiency of advertisements is low.

In order to solve the above problems, this application discloses a method for searching advertisement information, including:

Expanding the query words input by the client to obtain an expanded set of candidate words;

Analyze search engine real-time logs and/or advertiser status information at the current time to obtain an effective word set; the advertiser status information includes keyword display condition information selected by the advertiser; the effective word set includes keywords that meet the display conditions at the current time word;

Using the effective word set to filter out words that are not in the effective word set in the candidate word set;

At least one candidate word in the filtered candidate word set is selected to retrieve advertisement information.

Preferably, the user status information is obtained through the following steps:

The database table corresponding to each advertiser is obtained from the data storage server, and the database table includes the advertiser's advertising expense budget and the advertiser's advertisement promotion time.

Preferably, the database table includes an advertiser's advertising expense budget table and an advertiser's advertising promotion schedule, corresponding to the advertiser's advertising expense budget and advertising promotion time respectively.

Preferably, after acquiring the database table corresponding to the advertiser, it includes:

Analyzing the advertiser's advertising expense budget table and the advertiser's advertising promotion timetable, for the advertiser whose advertising expense budget is greater than the threshold and whose advertising promotion period includes the current time, add the keywords selected by the advertisement into effective words gather.

Preferably, the effective word set is obtained by analyzing the real-time log of the search engine through the following steps:

Get real-time search logs from search engines;

Analyze whether each retrieval keyword in the real-time search log returns an advertisement to the client within the current time period;

If it returns, add the corresponding search keyword to the effective word set.

Preferably, after using the effective word set to filter out the words in the candidate word set that are not in the effective word set, it also includes:

For the candidate words in the filtered candidate word set, calculate the correlation between the candidate words and the query word;

For the candidate words in the filtered candidate word set, the effective advertiser depth of the candidate words is counted; the effective advertiser depth is the advertiser who selects the candidate word and meets the display conditions at the current time quantity;

The candidate words in the filtered candidate word set are reordered in combination with the weight of the relevance degree of the candidate words and the weight of the effective advertiser depth.

Preferably, the database table also includes an advertiser online status table.

Preferably, after using the effective word set to filter out the words in the candidate word set that are not in the effective word set, it also includes:

For the candidate words in the filtered candidate word set, calculate the correlation between the candidate words and the query word;

For the candidate words in the filtered candidate word set, the effective advertiser depth of the candidate words is counted; the effective advertiser depth is the advertiser who selects the candidate word and meets the display conditions at the current time quantity;

Counting the number of online advertisers who select the candidate word and meet the display conditions at the current time;

The candidate words in the filtered candidate word set are reordered in combination with the weight of the relevance degree of the candidate words, the weight of the effective advertiser depth and the number of online advertisers.

Preferably, the query word input by the user terminal is expanded, and the extended candidate word set includes:

Obtaining the keywords newly selected by the advertiser, and adding the keywords into the expanded thesaurus;

The correlation between the query word and each keyword in the expanded thesaurus is calculated, and when the correlation is greater than a threshold, the keyword is added to the candidate word set.

Correspondingly, this application discloses an advertisement information search system, including:

The query word expansion module is used to expand the query words input by the client to obtain an expanded set of candidate words;

The effective word set acquisition module is used to analyze the search engine real-time log and/or the advertiser status information of the current time to obtain the effective word set; the advertiser status information includes the keyword display condition information selected by the advertiser; the effective word set Include keywords that meet the display conditions at the current time;

A filtering module, configured to use the effective word set to filter out words in the candidate word set that are not in the effective word set;

The retrieval module is used to select at least one candidate word in the filtered candidate word set to retrieve advertisement information.

Preferably, the effective word set acquisition module includes:

The user status information acquisition sub-module is used to acquire the database table corresponding to each advertiser from the data storage server, and the database table includes the advertiser's advertising expense budget and the advertiser's advertising promotion time.

Preferably, the database table includes an advertiser's advertising expense budget table and an advertiser's advertising promotion schedule, corresponding to the advertiser's advertising expense budget and advertising promotion time respectively.

Preferably, the effective word set acquisition module also includes:

The user status analysis sub-module is used to analyze the advertiser's advertising expense budget table and the advertiser's advertising promotion schedule, and for the advertiser whose advertising expense budget is greater than a threshold and whose advertising promotion time period includes the current time, the The keywords selected by the advertisement are added to the effective word set.

Preferably, the effective word set acquisition module includes:

The real-time log acquisition sub-module is used to obtain real-time search logs from search engines;

The real-time log analysis sub-module is used to analyze whether each search keyword in the real-time search log returns an advertisement to the client within the current time period; if returned, the corresponding search keyword is added to the effective word set.

Compared with the prior art, the present application includes the following advantages:

After the system obtains the query words at the user end, and obtains a large number of candidate word sets related to the query words according to the expansion of the query words, the application analyzes the real-time status data of the advertiser or the real-time logs of the search engine in real time, and obtains the currently recallable advertisements. keywords, and add the keyword as an effective keyword to the effective word set, then use the effective word set to screen the candidate word set, obtain a candidate word set capable of recalling the advertisement after screening, and then perform the screening on the Each candidate word is re-ranked to obtain at least one search keyword that ranks first among them, and at most the search keywords that the search engine maintains the optimal number of performance, and then search based on the above-mentioned search keywords. In the above method, by filtering out the extended candidate words that cannot recall advertisements in real time, it is ensured that the search keywords input to the search engine at the current time can recall advertisements for display in time, without affecting the display efficiency of advertisements, and making full use of the limited display Resource display ads.

Description of drawings

FIG. 1 is a schematic flow diagram of an advertisement information search method in the present application;

Fig. 2 is a schematic structural diagram of an advertisement information search system of the present application;

Fig. 3 is a schematic structural diagram of a preferred advertisement information search system of the present application.

detailed description

In order to make the above objects, features and advantages of the present application more obvious and comprehensible, the present application will be further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

Referring to FIG. 1 , it shows a schematic flow chart of an advertisement information search method of the present application, which may specifically include:

Step 110, expand the query word input by the client to obtain an expanded set of candidate words.

Due to the asymmetry between the knowledge fields of users and advertisers, sometimes the keywords selected by advertisers (that is, the bidding words purchased) may not completely conform to the user's habit of forming words when searching, resulting in the inability to maximize possible exposure. Promoted products or advertisements, so search engines generally expand the query words entered by users. Based on the keyword advertising index method generally used by search engines (matching corresponding advertisements through the bidding words purchased by users), search engine keyword advertising In extended matching, one or more bid words are generally expanded on the basis of the original user query through various methods (word addition, word subtraction, replacement, etc.) under the premise of ensuring a certain relevance, and then the original Words and the expanded bidding words are used as new queries for advertisement matching.

In keyword advertising, in addition to the relevance in the bidding word expansion matching, evaluation methods related to commercial factors are generally added, such as expanding the words with high PPC (pay-per-click, which means the deduction fee for each click of the advertisement), and the customer Words with high depth (the number of users who buy a certain bidding word is called customer depth) and so on.

Generally, there are three ways to expand query words: one is phrase matching, which takes substrings of user query words as recommended words according to relevance, such as Nokia mobile phones, which can expand Nokia and mobile phones, among which Nokia is more relevant than mobile phones.

The second is to include extended matching, that is, words with richer extended information include words searched by users. For example, when users search for mobile phones, Samsung mobile phones, htc mobile phones, etc. can be extended. Index to find all the words that contain the original word, and finally select the best N words according to the relevance as the recommendation result.

The third is related recommendation based on statistics. According to the search log of the website, search for similar search words according to user search behavior, perform clustering, and recommend related words. For example, searching for mobile phones can recommend iphone.

This application also expands the query words input by the user terminal, and first obtains the expanded candidate word set.

Step 120, analyze the search engine real-time log and/or advertiser status information at the current time to obtain an effective word set; the advertiser status information includes the keyword display condition information selected by the advertiser; the effective word set includes the current time to meet the display condition information Condition keyword.

Because the search engine has a limit on the maximum number of search terms based on its performance, it will only select the most relevant search terms. Among the extended words obtained by the prior art, the keywords that may have been selected by the advertiser (that is, the bidding word selected by the client), because the advertiser has set the display conditions for the advertisements that need to be promoted, so the advertiser chooses The keywords also have corresponding display conditions. If the current display environment of the search engine does not meet the display conditions set by the user, for example, the ad display time period set by the customer does not include the current display time, or the customer's current account balance is lower than its setting Even if the search engine finally selects the keyword, it cannot recall the advertisement of the advertiser corresponding to the keyword for display, which wastes the search work of the search engine and reduces the efficiency of the search engine.

In this application, for the candidate word set obtained above, first obtain a real-time effective word set, which includes keywords that meet the display conditions at the current time, for example, for "Nokia", there is an advertiser who selects this word, The advertisement display time period set by it includes the current display time, and at the same time, the account balance used to pay the promotion fee is greater than the threshold value set by it. Among them, general promotion in search engines generally adopts CPC (Cost Per Click, cost per click for online advertising) system to collect fees from advertisers.

For obtaining the effective word set, the effective word set is generally obtained by analyzing the real-time log of the search engine and/or the advertiser status information at the current time.

Preferably, the user status information is obtained through the following steps:

In step S11, the database table corresponding to each advertiser is obtained from the data storage server, and the database table includes the advertiser's advertising expense budget and the advertiser's advertising promotion time.

Further, the method includes: acquiring a database table corresponding to each advertiser from the data storage server, and the database table includes the advertiser's advertising expense budget table and the advertiser's advertising promotion schedule.

This application is associated with the database table. Generally, there are two factors that determine whether an advertiser's advertisement is online. One is the budget, that is, the advertiser's setting of the limit value of his own account balance. For example, when the account balance is less than 300, no Then allow the advertisement to be displayed; the second is the promotion time, that is, the advertiser's setting of the advertisement promotion time, such as 11:00 noon to 1:00 pm. Therefore, the database table corresponding to each advertiser can be obtained from the data storage server for analysis. Furthermore, the effective word set can be obtained through the advertiser's advertising expense budget table and the advertiser's advertising promotion timetable in the database table, that is, to obtain Effective Advertiser Depth (Effective Advertiser Depth: For each advertiser who purchased a bidding word, at a certain moment, the number of advertisers corresponding to the advertisements that can be displayed online through the bidding word, that is, at the certain moment, The number of advertisers (number of advertisers corresponding to the advertisements to be displayed) can be recalled from the advertisement library through the bidding words (bidding words).

Preferably, after acquiring the database table corresponding to the advertiser, it includes:

Step S12, analyzing the advertiser's advertising expense budget table and the advertiser's advertising promotion schedule, and for the advertiser whose advertising expense budget is greater than a threshold and whose advertising promotion period includes the current time, the key selected by the advertiser Words are added to the effective word set.

That is, through the correlation between the advertiser's advertising expense budget table and the advertiser's advertising promotion schedule and the keywords selected by the advertiser, the effective word set is obtained through analysis.

Or, analyze the search engine real-time log to obtain the effective word set through the following steps:

Step S21, obtaining a real-time search log from a search engine;

Step S22, analyzing whether each retrieval keyword in the real-time search log returns an advertisement to the client within the current time period;

Step S23, if it returns, add the corresponding search keyword to the effective word set.

In the search engine, the user's search query words and the advertisements displayed by the search engine will be recorded in real time, so through this method, the number of real-time returned advertisements for the user's query words within this period of time can be obtained, that is, the real-time customer depth. In , if the advertisements are returned in real time through the query words and their expansion words, then these words can be considered as effective words and added to the effective word set.

Step 130, using the effective word set to filter out words in the candidate word set that are not in the effective word set.

After the effective word set is obtained, the candidate word set can be processed to filter out the words in the candidate word set that are not in the effective word set. In this way, words without customer depth can be filtered out, that is, words that cannot recall an advertiser's advertisement for display. For example, if the current time is 14:00, for the keyword "Nokia", there are advertisers A, B, and C who purchased this word:

For A, the advertisement display time corresponding to the keyword set by it is 11:00-13:00, the account balance is 1000, and the account balance threshold set by it is 300;

For B, the advertisement display time corresponding to the keyword set by it is 11:00-15:00, the account balance is 100, and the account balance threshold set by it is 400;

For C, the advertisement display time corresponding to the keyword set by it is 18:00-21:00, the account balance is 200, and the account balance threshold set by it is 300;

Then, for the keyword "Nokia", for A, although its account balance is greater than the threshold of 300 set by A in the display conditions, the current display time does not meet the requirements of A, so the keyword "Nokia" cannot recall the corresponding "Nokia" of A. ” ads are displayed;

For B, although the current display time meets B's requirements in its display conditions, its account balance is less than the threshold set by B of 400, so the keyword "Nokia" cannot recall B's corresponding "Nokia" advertisement for display;

For C, the current display time point in its display conditions does not meet the requirements of C, and its account balance is less than the threshold value of 300 set by C, then the keyword "Nokia" cannot recall B's corresponding "Nokia" advertisement for display.

In addition, preferably, after using the effective word set to filter out the words in the candidate word set that are not in the effective word set, it also includes:

Step 131, for the candidate words in the filtered candidate word set, calculate the correlation between the candidate words and the query word;

For the filtered candidate words, the correlation between the candidate words and the query word can be calculated, such as literal matching text similarity + topic similarity:

Literal text similarity For example: "Nokia mobile phone" and "Samsung mobile phone", first segment the two words and label them, "Nokia/brand", "mobile phone/product", "Samsung/brand", "mobile phone/product" , the importance of different tags is different, for example, the product tag is greater than the brand tag, greater than the modifier tag, and then calculate the similarity score based on the word segmentation and its tag matched by the two words.

For example, based on the theme system of the website itself (such as digital products, hardware tools, sports, entertainment), by calculating the information clicked by the user after searching the query word, the topic tendency of each query word can be calculated, and each query word There will be one or more topics, and the similarity between the two is calculated according to the number of matching topics.

Step 132, for the candidate words in the filtered candidate word set, count the effective advertiser depth of the candidate words; the effective advertiser depth is the selection of the candidate word and meet the display conditions at the current time the number of advertisers;

For the filtered candidate words, because they are filtered by the effective word set, each candidate word will have an effective advertiser depth, that is, the number of advertisements that can be recalled and displayed by the word at the current time. For example, for "laptop computer", the current time is 12:00, the customers who purchased this word include M1, M2, M3, M4, M5; 12:00, and the account balances of M1, M2, M3, and M4 are greater than the minimum threshold set by them, and the display time period of the advertisement corresponding to the word set by M5 does not include 12:00, then the effective advertisement of the word "laptop" The customer depth is 4.

Step 133, combining the weight of the relevance of the candidate words and the weight of the effective advertiser depth to reorder the candidate words in the filtered candidate word set.

In practice, when sorting, weights will be set for the relevance of the candidate word and the query word, and the weight of the effective advertiser depth of the candidate word will also be set, so the weight of the relevance degree of the candidate word and the effective advertiser depth of the candidate word Weight calculates the total weight of the candidate words, so that the ranking can be based on the total weight of each candidate word. For example, the relevance weight is a, the weight of the effective advertiser depth is b, for the candidate word P, its correlation with the query word is m, and its effective advertiser depth is n, then the total weight of the word P is W=a* m+b*n.

In addition, the database table also includes an advertiser online status table. Used to indicate user online status. Preferably, after using the effective word set to filter out the words in the candidate word set that are not in the effective word set, it also includes:

Step S31, calculating the correlation between the candidate words and the query word for the candidate words in the filtered candidate word set;

Step S32, for the candidate words in the filtered candidate word set, calculate the effective advertiser depth of the candidate words; the effective advertiser depth is the selected candidate word and meets the display conditions at the current time the number of advertisers;

Step S33, counting the number of online advertisers who select the candidate word and meet the display conditions at the current time;

Step S34 , reordering the candidate words in the filtered candidate word set in combination with the weight of the relevance of the candidate words, the weight of the effective advertiser depth and the number of online advertisers.

This application can also set a weight for the number of online advertisers of the candidate word, and add it to the calculation of the total weight.

Step 140, selecting at least one candidate word in the filtered candidate word set to retrieve advertisement information.

Preferably, among the aforementioned sorted candidate words, at least one candidate word ranked higher is selected to retrieve advertisement information in a search engine.

Referring to Fig. 2, it shows a schematic structural diagram of an advertisement information search system of the present application, including:

The query word expansion module 210 is used to expand the query word input by the client to obtain an expanded set of candidate words;

Effective word set acquisition module 220, used to analyze search engine real-time log and/or the advertiser state information of current time to obtain effective word set; Described advertiser state information comprises the keyword display condition information that advertiser selects; Described effective word The collection includes keywords that meet the display conditions at the current time;

A filtering module 230, configured to use the effective word set to filter out words in the candidate word set that are not in the effective word set;

The retrieval module 240 is configured to select at least one candidate word in the filtered candidate word set to retrieve advertisement information.

Preferably, the effective word set acquisition module includes:

The user status information acquisition sub-module is used to acquire the database table corresponding to each advertiser from the data storage server, and the database table includes the advertiser's advertising expense budget and the advertiser's advertising promotion time.

The database table includes an advertiser's advertising expense budget table and an advertiser's advertising promotion schedule, which correspond to the advertiser's advertising expense budget and advertisement promotion time respectively.

The effective word collection acquisition module also includes:

The user status analysis sub-module is used to analyze the advertiser's advertising expense budget table and the advertiser's advertising promotion schedule, and for the advertiser whose advertising expense budget is greater than a threshold and whose advertising promotion time period includes the current time, the The keywords selected by the advertisement are added to the effective word set.

The effective word collection acquisition module includes:

The real-time log acquisition sub-module is used to obtain real-time search logs from search engines;

The real-time log analysis sub-module is used to analyze whether each search keyword in the real-time search log returns an advertisement to the client within the current time period; if returned, the corresponding search keyword is added to the effective word set.

Referring to Fig. 3, it shows a schematic structural diagram of a preferred advertisement information search system of the present application, including:

User query word processing server 310, search engine server 320, data storage server 330;

The user query word processing server 310 is used to receive the user's original query word obtained by the front end of the search engine server 320, and obtain the advertiser status information stored in the data storage server 330; and/or the search engine server The real-time search log at 320 processes the query words, and returns the processed words to the search engine server 320 .

The user query word processing server 310 includes: a query word expansion module, which is used to expand the query word input by the user terminal to obtain an expanded set of candidate words;

The valid word set acquisition module is used to obtain the effective word set for the advertiser status information of the current time; the advertiser status information includes the keyword display condition information selected by the advertiser; the effective word set includes the keywords that meet the display conditions at the current time word;

A filtering module, configured to use the effective word set to filter out words in the candidate word set that are not in the effective word set.

The search engine server 320 is configured to input the original query words input by the user into the user query word processing server 310 , and perform final advertisement retrieval based on the search words sent by the user query word processing server 310 . Wherein, the search engine server includes a retrieval module for selecting at least one candidate word in the filtered candidate word set to retrieve advertisement information.

The data storage server 330 is used for storing advertiser status information.

As for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for the related parts, please refer to the part of the description of the method embodiment.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.

The present application is described with reference to flowcharts and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram. These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, thereby The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

While preferred embodiments of the present application have been described, additional changes and modifications to these embodiments can be made by those skilled in the art once the basic inventive concept is appreciated. Therefore, the appended claims are intended to be construed to cover the preferred embodiment and all changes and modifications which fall within the scope of the application.

The above is a detailed introduction to the advertising information search method and system provided by this application. In this paper, specific examples are used to illustrate the principle and implementation of this application. The description of the above embodiment is only used to help understand this application. method and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation and scope of application. In summary, the content of this specification should not be understood as Limitations on this Application.

Claims (14)

1. A kind of 1. advertising message searching method, it is characterised in that including：

   The query word that user terminal inputs is extended, the candidate word set after being extended；

   The advertiser's status information for analyzing search engine real-time logs and/or current time obtains effective set of words；It is described wide Accusing client state information includes the keyword displaying conditional information of advertiser's selection；The effectively set of words includes current time Meet the keyword of displaying condition；

   Using effective set of words, the not word in effective set of words is filtered out in the candidate word set；Wherein, it is described not Word in effective set of words refers to that the word that the advertisement of advertiser is showed can not be recalled；

   At least one candidate's word and search advertising message in candidate word set after selection filtering.
2. 2. according to the method for claim 1, it is characterised in that obtain user state information by following steps：

   Database table corresponding to each advertiser is obtained from data storage server, the database table includes the wide of advertiser Accuse expense budget and the advertisement promotion time of advertiser.
3. 3. according to the method for claim 2, it is characterised in that：

   The database table includes the advertising expenditure budget table of advertiser and the advertisement promotion timetable of advertiser, right respectively Answer advertising expenditure budget and the advertisement promotion time of advertiser.
4. 4. according to the method for claim 3, it is characterised in that after obtaining database table corresponding to advertiser, including：

   The advertising expenditure budget table of the advertiser and the advertisement promotion timetable of advertiser are analyzed, it is pre- for advertising expenditure The advertiser that calculation is more than threshold value and the advertisement promotion period includes the current time, the keyword of the advertisement selection is added Enter effective set of words.
5. 5. according to the method for claim 1, it is characterised in that analyze search engine real-time logs by following steps and obtain Effective set of words：

   Search daily record in real time is obtained from search engine；

   Analyze whether each search key in the daily record of search in real time returns to advertisement to user terminal in current slot；

   If returned, corresponding search key is added into effective set of words.
6. 6. according to the method for claim 1, it is characterised in that utilizing effective set of words, filtering out the candidate In set of words not after the word in effective set of words, in addition to：

   For the candidate word in the candidate word set after the filtering, the degree of correlation of the candidate word and the query word is calculated；

   For the candidate word in the candidate word set after the filtering, effective advertiser's depth of the candidate word is counted；Institute The advertiser's for stating effective advertiser's depth to select the candidate word and meet the displaying condition in current time Quantity；

   With reference to the weight of the degree of correlation of the candidate word and the weight of effective advertiser's depth to the candidate word after the filtering Each candidate word in set is reordered.
7. 7. according to one of them the described method of claim 3,4, it is characterised in that：

   The database table also includes advertiser's presence table.
8. 8. according to the method for claim 1, it is characterised in that utilizing effective set of words, filtering out the candidate In set of words not after the word in effective set of words, in addition to：

   For the candidate word in the candidate word set after the filtering, the degree of correlation of the candidate word and the query word is calculated；

   For the candidate word in the candidate word set after the filtering, effective advertiser's depth of the candidate word is counted；Institute The advertiser's for stating effective advertiser's depth to select the candidate word and meet the displaying condition in current time Quantity；

   Statistics selects the candidate word and the quantity for the online advertisement client for meeting the displaying condition in current time；

   The weight and online advertisement customer quantity of weight, effective advertiser's depth with reference to the degree of correlation of the candidate word are to institute Each candidate word stated in the candidate word set after filtering is reordered.
9. 9. according to the method for claim 1, it is characterised in that the query word that user terminal inputs is extended, expanded Include during candidate word set after exhibition：

   The keyword that advertiser newly selects is obtained, the keyword is added into extension dictionary；

   The query word is calculated with extending the correlation of each keyword in dictionary, when the correlation is more than threshold value, by described in Keyword adds candidate word set.
10. A kind of 10. advertising message search system, it is characterised in that including：

    Query word expansion module, the query word for user terminal to be inputted are extended, the candidate word set after being extended；

    Effective set of words acquisition module, advertiser's state for analyzing search engine real-time logs and/or current time are believed Breath obtains effective set of words；Advertiser's status information includes the keyword displaying conditional information of advertiser's selection；Institute Stating effective set of words includes the keyword that current time meets displaying condition；

    Filtering module, for utilizing effective set of words, filter out in the candidate word set not in effective set of words Word；Wherein, the word not in effective set of words refers to that the word that the advertisement of advertiser is showed can not be recalled；

    Module is retrieved, for selecting at least one candidate's word and search advertising message in the candidate word set after filtering.
11. 11. system according to claim 10, it is characterised in that the effectively set of words acquisition module includes：

    User state information acquisition submodule, for obtaining database table corresponding to each advertiser from data storage server, The database table includes the advertising expenditure budget of advertiser and the advertisement promotion time of advertiser.
12. 12. system according to claim 11, it is characterised in that：

    The database table includes the advertising expenditure budget table of advertiser and the advertisement promotion timetable of advertiser, right respectively Answer advertising expenditure budget and the advertisement promotion time of advertiser.
13. 13. system according to claim 12, it is characterised in that the effectively set of words acquisition module also includes：

    User Status analyzes submodule, is pushed away for analyzing the advertisement of advertising expenditure budget table and advertiser of the advertiser Wide timetable, it is more than the advertiser that threshold value and advertisement promotion period include the current time for advertising expenditure budget, The keyword of the advertisement selection is added into effective set of words.
14. 14. system according to claim 10, it is characterised in that the effectively set of words acquisition module includes：

    Real-time logs acquisition submodule, for obtaining search daily record in real time from search engine；

    Real-time logs analyze submodule, are in current slot for analyzing each search key in the daily record of search in real time No return advertisement is to user terminal；If returned, corresponding search key is added into effective set of words.