CN103136310A - Method for conversion of new queries based on historical queries of specific websites - Google Patents
Method for conversion of new queries based on historical queries of specific websites Download PDFInfo
- Publication number
- CN103136310A CN103136310A CN2011104138267A CN201110413826A CN103136310A CN 103136310 A CN103136310 A CN 103136310A CN 2011104138267 A CN2011104138267 A CN 2011104138267A CN 201110413826 A CN201110413826 A CN 201110413826A CN 103136310 A CN103136310 A CN 103136310A
- Authority
- CN
- China
- Prior art keywords
- query
- website
- inquiry
- candidate
- original
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention discloses a method for conversion of new queries based on historical queries of specific websites. Each specific website has respective content preference and respective user groups. When a user inquires on a certain website, the query is related to the content of the website on most conditions. By utilization of historical search information of the website, the original query which is possibly ambiguous is converted. Then, based on the substitute query, a search is carried out with an advertisement engine, and advertisements searched are returned to the user, so that a query item which is really used for advertisement searching is more matched with a target task of the website, and meanwhile the advertisement result is more matched with the search intent of the user.
Description
Technical field
The present invention relates to the advertisement search technology, more particularly, relate to the technology of according to the information of specific website, original query being changed in website alliance (syndication network).
Background technology
Commercial advertisement engine (such as bing, google) provides the advertisement support usually for a lot of other websites, and these websites have the page, search inputting interface and search engine separately, but when carrying out the backstage search, use be same advertisement engine.When some websites was searched for, two class results are returned in this website: a class was Search Results as the user, was provided or third party's search engine provides by website self; Another kind of is the advertisement result, is provided by the advertisement engine of website alliance.
Different websites has different theme tendency and target service, the fancy grade of the advertisement that therefore advertisement engine is provided is not identical yet, the website can wish that the relevant advertisement of target service that will pay close attention to the most to oneself is arranged in forward position, these advertisements are more likely clicked by the user like this, and user's click can bring income to the website.The quantity of the advertisement that the target service that click is paid close attention to the most to oneself is relevant is more, and the income that the website obtains is also just higher.
In the alliance of existing website, the advertisement result that each website provides the advertisement engine that shares has been carried out certain processing according to the target service of oneself.Such as, website " jobs.Yahoo.com " share advertisement engine with " bing ", but both target service are different.The target service of " jobs.yahoo.com " is working opportunity and job ad, and the target service of " bing " is universal search engine.So if a same key word is searched on the webpage of this website, the Search Results that obtains is different, corresponding, the advertisement result also should be different.
Such as, input " computer " is searched in " jobs.yahoo.com " and " bing " equally, obtains different Search Results.
" jobs.yahoo.com " is all relevant to job overall for arranging forward Search Results in the Search Results of " computer ", and this is because " jobs.yahoo.com " self characteristic causes.Because " jobs.yahoo.com " is a website take job overall as target service, the Search Results relevant with job overall and advertisement result are seen in user's expectation, they only can click the webpage relevant with job overall or advertisement, and only have advertisement to bring income for " jobs.yahoo.com "; The user can not click the advertisement irrelevant with job overall, therefore can not bring income for " jobs.yahoo.com ".
" bing " occupied the forward position of Search Results for the webpage of selling relevant for computer product in the Search Results of " computer ".The target service of " bing " is universal search engine, its revenue source professional website such from being similar to " jobs.yahoo.com " is different, therefore " bing " can select the advertisement result the most favourable to its income, and these advertisement results are very different with the result of " jobs.yahoo.com ".
For the website in the website alliance that uses the advertisement engine that shares, need to select to meet the advertisement of own target service from the advertisement result of advertisement engine feedback, could effectively improve clicking rate like this, improve income.Although had some to process to the advertisement result at present, these processing are not to have very high efficient and good applicability.In the search of some key words, these processing can reach certain effect, but in the processing of considerable key word, present scheme can't reach effective processing, for specific website, still can will be arranged in forward position in the irrelevant advertisement of self target service on its webpage.
Summary of the invention
Thereby the present invention is intended to propose a kind ofly can provide based on the method that the historical query of the specific website in the alliance of website is changed new inquiry the method for advertisement search result more accurately.
In one embodiment, the present invention has disclosed a kind of method of new inquiry being changed based on the historical query of individual Web sites.The method at first other website one by one in the alliance of website receives user's original query, and this website alliance comprises several independently website and Topic relatives of this original query and this website.The method is obtained all users' historical search information from this website afterwards, based on this historical search information, described original query is changed, and obtains the replacement query through conversion.Search in advertisement engine based on this replacement query at last, the advertisement that searches is shown to the user.
In one embodiment, the present invention has disclosed a kind of method of new inquiry being changed based on the historical query of individual Web sites.The method at first other website one by one in the alliance of website receives original query, and this website alliance comprises several independently website and Topic relatives of this original query and this website.Then the method is obtained all users tendency query term relevant with original query in a period of time in the past from this website.Should be inclined to query term and merge in original query, obtain the replacement query through conversion.At last search in advertisement engine and the advertisement that searches is shown to the user based on this replacement query.
In one embodiment, the present invention has disclosed a kind of method of new inquiry being changed based on the historical query of individual Web sites.The method at first other website one by one in the alliance of website receives original query, and this website alliance comprises several independently website and Topic relatives of this original query and this website.Then obtain candidate inquiry from this website and candidate's inquiry is screened, this screening is based on candidate's inquiry and the attribute of original query and compares with the frequency of utilization of similarity and candidate's inquiry and carry out, and screening obtains the candidate's inquiry with the original query coupling.Use the candidate of this coupling to inquire about the described original query of replacement, obtain the replacement query through conversion.At last search in advertisement engine and the advertisement that searches is shown to the user based on this replacement query.
The present invention can change query term pointedly for characteristic and the historical data of specific website in the alliance of website, make real query term for advertisement search and the target service of website more mate, thereby obtain the valuable advertisement search result of tool more.
Description of drawings
The above and other features of the present invention, character and advantage will become more obvious by the description below in conjunction with drawings and Examples, in the accompanying drawings, identical Reference numeral represents identical feature all the time, wherein:
Fig. 1 has disclosed the process flow diagram according to the method for new inquiry being changed based on the historical query of individual Web sites of the first embodiment of the present invention.
Fig. 2 has disclosed the process flow diagram of the method for new inquiry being changed based on the historical query of individual Web sites according to a second embodiment of the present invention.
Fig. 3 has disclosed the process flow diagram of the method for new inquiry being changed based on the historical query of individual Web sites of a third embodiment in accordance with the invention.
Embodiment
The present invention proposes a kind of method of new inquiry being changed based on the historical query of individual Web sites.Fig. 1 has disclosed the process flow diagram according to the method for the first embodiment of the present invention.Main thought of the present invention is: a specific website in the alliance of website receives an original query, log on the webpage of this website such as the user and after having inputted an inquiry, this website is analyzed, obtained the information relevant to its target service from this website.These information will be used to original query is changed, and the process of conversion is that the proportion relevant to the target service of this website is increased, and obtains an inquiry of more mating with the target service of website.When using advertisement engine to search for, with the query term that uses after changing, the Search Results that obtains like this is more relevant to the target service of website.
The method is applicable to have any website that the limited occupation feature is arranged.With reference to shown in Figure 1, the method 100 comprises:
102. the other website one by one in the alliance of website receives user's original query.This website alliance comprises several independently, have the website of service feature separately, and this original query is relevant to this specific website.An implementation of step 102 is: user's input inquiry on specific website is searched for, and has just thought to input an original query relevant to this website.The explanation of giving one example, the user logs on " jobs.yahoo.com " and has inputted " computer ", thinks that the user has inputted an original query relevant to " jobs.yahoo.com ", and this original query is " computer ".
104. obtain all users' historical search information from this website, based on this historical search information, described original query is changed, obtain the replacement query through conversion.Historical search information is relevant to this specific website, usually comes from the historical data of this specific website, such as the once used historical query of this website, the historical search result that obtains, history web pages of browsing etc.This historical search information has the form of two kinds of uses: a kind of mode is that historical search information is merged in original query, obtains the replacement query through conversion, and in this mode, historical search information is a tendency query term.The second embodiment relates to this mode.Another kind of mode is to replace original query with historical search information, inquiry as an alternative, and in this mode, historical search information is also an inquiry, the 3rd embodiment relates to this mode.
106. search in advertisement engine based on this replacement query, the advertisement that searches be shown to the user.
Fig. 2 has disclosed method according to a second embodiment of the present invention, and in the method 200, historical search information is the form of tendency query term, and the tendency query term is integrated in original query.As shown in Figure 2, the method 200 comprises:
202. other website one by one in the alliance of website receives original query.This website alliance comprises several independently websites, the Topic relative of this original query and this website.Step 202 is similar with step 102, no longer is repeated in this description herein.
204. obtain all users tendency query term relevant with original query in a period of time in the past from this website.The obtain manner of tendency query term has following two kinds:
1) obtain the tendency query term from the historical query of this specific website.Such as searching for the highest query term of the frequency of occurrences as the tendency query term in the historical query of this specific website.Refer again to the front for the example in " jobs.yahoo.com " upper input " computer ", the tendency query term is from the website, namely analyze in " jobs.yahoo.com " and obtain, analyze the historical query record of " jobs.yahoo.com ", namely all are at the record of the inquiry of " jobs.yahoo.com " upper input, find the wherein the highest query term of the frequency of occurrences, key word is " job " in other words, so just " job " is chosen as the tendency query term.
2) obtain the tendency query term from the webpage of this specific site search.Such as at the highest query term of the netpage search frequency of occurrences of this specific site search as the tendency query term.Or with reference to for the example at " jobs.yahoo.com ", the Search Results that the inquiry of being undertaken by " jobs.yahoo.com " is searched, webpage is analyzed in other words, search the highest query term of frequency that occurred on these webpages, discovery is " job ", so " job " just is selected as the tendency query term.
Merged in original query 206. should be inclined to query term, obtained the replacement query through conversion.For the tendency query term, it is considered to be a kind of extra querying condition for original query, to improve the accuracy rate of inquiry.The tendency query term will merge and obtain replacement query in original query.Such as, in front for example in, the original query " computer " of inputting by " jobs.yahoo.com " and tendency query term " job " merge the replacement query " computer+job " after being changed.
208. search in advertisement engine based on this replacement query, the advertisement that searches be shown to the user.
Fig. 3 has disclosed the method for a third embodiment in accordance with the invention, and in the method 300, historical search information is also one and inquires about and be used to replace original query.As shown in Figure 3, the method 300 comprises:
302. other website one by one in the alliance of website receives original query.This website alliance comprises several independently websites, the Topic relative of this original query and this website.Step 302 is similar with step 102 and step 202, no longer is repeated in this description herein.
304. obtain candidate's inquiry from this website.Candidate's inquiry is to obtain from the historical query of the website of this appointment, and in order to ensure the relevance of candidate's inquiry with original query, candidate's inquiry has at least one identical query term with original query.Again take the inquiry " computer " inputted by " jobs.yahoo.com " as example, original query is " computer ", in the historical query of " jobs.yahoo.com ", select to have at least one identical query term with original query, i.e. all inquiries that comprise query term " computer " are inquired about as the candidate.Herein, a byte during query term refers to inquire about, perhaps, for text query, a query term refers to a word.More specifically, query term is word in Chinese or a word in English.
306. candidate inquiry is screened, and this screening is based on candidate's inquiry and the attribute of original query and compares with the frequency of utilization of similarity and candidate's inquiry and carry out, screening obtains the candidate's inquiry with the original query coupling.Comprise the consideration of three aspects for the screening of candidate's inquiry:
1) degree that is consistent on attribute with original query;
2) with the similarity of original query;
3) frequency of utilization of candidate's inquiry.
For the consideration of above-mentioned three aspects, following restrictive condition has been proposed:
Article one, restrictive condition is to compare for candidate's inquiry and the attribute of original query, candidate's inquiry all is comprised of a group polling item with original query, each query term is considered to a byte (term), for text query, a query term is exactly a word, such as for " computer device " this inquiry, thinks that it has two query terms (two bytes), be respectively " computer " and " device ", the byte length of this inquiry is 2.Article one, restrictive condition requires to have between candidate's inquiry and original query and is no more than the difference of following one:
Candidate inquiry and original query have identical byte (term) length and a different query term;
The second restrictive condition is compared for the similarity of candidate inquiry and original query, and this comparison is according to being inverted text frequency (IDF) and carrying out.Particularly, comprising:
The inversion text frequency (IDF) of calculated candidate inquiry and original query.Text frequency (DF) for from the given query term of given website refers to the search rate at a period of time inner cap query term, can be calculated as follows and be inverted text frequency (IDF):
Wherein maxDF has the search rate of the query term of high search rate.
Then based on being inverted the similarity of text frequency computation part candidate inquiry with original query.If q
iAnd q
jBe two inquiries, such as original query and candidate's inquiry, inquire about q for these two
iAnd q
jBetween similarity be calculated as:
S wherein
1And S
2Be respectively the inversion text frequency of original query and candidate inquiry.
Then screen similarity greater than candidate's inquiry of predetermined threshold, the similarity between second restrictive condition requirement original query and candidate's inquiry is higher than the threshold value of a setting, and this threshold value can be adjusted according to the requirement of using.
Article three, restrictive condition is the restrictive condition for the frequency of utilization of candidate's inquiry.Can predict, meet more than one of candidate's inquiry meeting of article one restrictive condition and second restrictive condition, so the 3rd restrictive condition limits the frequency of utilization of candidate's inquiry.Generally can select to have candidate's inquiry of the highest frequency of utilization in candidate's inquiry of the restrictive condition that meets article one and second.As a kind of realization, can select to have candidate's inquiry that peak hits rate.
Article one restrictive condition and second restrictive condition can guarantee that the candidate inquires about and original query between semantic similarity degree, in the situation that meet the characteristic that user's search request embodies the website as far as possible.And the 3rd restrictive condition guaranteed the frequency that is used that the candidate inquires about, and the high inquiry of frequency of utilization can access more Search Results usually, also can create better income.
308. use the candidate of this coupling to inquire about the replacement original query, obtain the replacement query through conversion.
310. search in advertisement engine based on this replacement query, the advertisement that searches be shown to the user.
The present invention can change query term pointedly for characteristic and the historical data of specific website in the alliance of website, make real query term for advertisement search and the target service of website more mate, thereby obtain the valuable advertisement search result of tool more.
Claims (16)
1. a method of new inquiry being changed based on the historical query of individual Web sites, is characterized in that, the method comprises:
Other website one by one in the alliance of website receives user's original query, and this website alliance comprises several independently websites, the Topic relative of this original query and this website;
Obtain all users' historical search information from this website, based on this historical search information, described original query is changed, obtain the replacement query through conversion;
Search in advertisement engine based on this replacement query, the advertisement that searches is shown to the user.
2. the method for claim 1, is characterized in that, described historical search information is relevant to described website.
3. method as claimed in claim 2, is characterized in that, described historical search information is integrated in described original query, obtains the replacement query through conversion.
4. method as claimed in claim 2, is characterized in that, described historical search information is used to replace described original query, as the replacement query through conversion.
5. a method of new inquiry being changed based on the historical query of individual Web sites, is characterized in that, the method comprises:
Other website one by one in the alliance of website receives original query, and this website alliance comprises several independently websites, the Topic relative of this original query and this website;
Obtain all users tendency query term relevant with original query in a period of time in the past from this website;
Should be inclined to query term and merge in described original query, obtain the replacement query through conversion;
Search in advertisement engine based on this replacement query, the advertisement that searches is shown to the user.
6. method as claimed in claim 5, is characterized in that, obtains the tendency query term and comprise obtain the tendency query term from the historical query of this website.
7. method as claimed in claim 6, is characterized in that, obtains the tendency query term and be included in query term that in the historical query of this website, the search frequency of occurrences is the highest as the tendency query term.
8. method as claimed in claim 5, is characterized in that, obtains the tendency query term and comprise obtain the tendency query term from the webpage of this site search.
9. method as claimed in claim 8, is characterized in that, obtains the tendency query term and be included in the highest query term of the netpage search frequency of occurrences of this site search as the tendency query term.
10. a method of new inquiry being changed based on the historical query of individual Web sites, is characterized in that, the method comprises:
Other website one by one in the alliance of website receives original query, and this website alliance comprises several independently websites, the Topic relative of this original query and this website;
Obtain candidate's inquiry from this website;
Candidate inquiry is screened, and this screening is based on candidate's inquiry and the attribute of original query and compares with the frequency of utilization of similarity and candidate's inquiry and carry out, and screening obtains the candidate's inquiry with the original query coupling;
Use the candidate of this coupling to inquire about the described original query of replacement, obtain the replacement query through conversion;
Search in advertisement engine based on this replacement query, the advertisement that searches is shown to the user.
11. method as claimed in claim 10 is characterized in that, described candidate's inquiry is to obtain from the historical query of this website.
12. method as claimed in claim 11 is characterized in that, described candidate's inquiry has at least one identical query term with original query.
13. method as claimed in claim 12 is characterized in that, described query term is word in Chinese or a word in English.
14. method as claimed in claim 11 is characterized in that, compares to screen based on candidate's inquiry and the attribute of original query to comprise that screening meets candidate's inquiry of one of following condition:
Candidate inquiry and original query have identical byte (term) length and a different query term;
Candidate's inquiry is lacked a byte than original query;
Candidate inquiry is Duoed a byte than original query.
15. method as claimed in claim 11 is characterized in that, screens with the similarity of original query based on candidate's inquiry to comprise:
The inversion text frequency (IDF) of calculated candidate inquiry and original query;
Based on being inverted the similarity of text frequency computation part candidate inquiry with original query;
The screening similarity is greater than candidate's inquiry of predetermined threshold.
16. method as claimed in claim 11 is characterized in that, screens based on the frequency of utilization of candidate inquiry to comprise:
Screening has candidate's inquiry that peak hits rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110413826.7A CN103136310B (en) | 2011-12-02 | 2011-12-02 | The method changed based on the historical query of individual Web sites to new inquiry |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110413826.7A CN103136310B (en) | 2011-12-02 | 2011-12-02 | The method changed based on the historical query of individual Web sites to new inquiry |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103136310A true CN103136310A (en) | 2013-06-05 |
| CN103136310B CN103136310B (en) | 2017-11-28 |
Family
ID=48496141
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110413826.7A Active CN103136310B (en) | 2011-12-02 | 2011-12-02 | The method changed based on the historical query of individual Web sites to new inquiry |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103136310B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108713186A (en) * | 2015-11-23 | 2018-10-26 | 微软技术许可有限责任公司 | It is prioritized locative search term |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1761972A (en) * | 2003-03-18 | 2006-04-19 | Nhn株式会社 | A method of determining an intention of internet user, and a method of advertising via internet by using the determining method and a system thereof |
| CN1877581A (en) * | 2006-07-12 | 2006-12-13 | 百度在线网络技术（北京）有限公司 | Advertisement display system and method used for Internet search engine |
| CN1961316A (en) * | 2004-05-29 | 2007-05-09 | Nhn株式会社 | Method and system for managing the impressing of the search listing based on advertisement group |
| CN102067105A (en) * | 2008-03-18 | 2011-05-18 | 雅虎公司 | Personalizing sponsored search advertising layout using user behavior history |
| CN102096882A (en) * | 2002-09-24 | 2011-06-15 | Google公司 | Methods and apparatus for serving relevant advertisements |
| US20110184936A1 (en) * | 2010-01-24 | 2011-07-28 | Microsoft Corporation | Dynamic community-based cache for mobile search |
-
2011
- 2011-12-02 CN CN201110413826.7A patent/CN103136310B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102096882A (en) * | 2002-09-24 | 2011-06-15 | Google公司 | Methods and apparatus for serving relevant advertisements |
| CN1761972A (en) * | 2003-03-18 | 2006-04-19 | Nhn株式会社 | A method of determining an intention of internet user, and a method of advertising via internet by using the determining method and a system thereof |
| CN1961316A (en) * | 2004-05-29 | 2007-05-09 | Nhn株式会社 | Method and system for managing the impressing of the search listing based on advertisement group |
| CN1877581A (en) * | 2006-07-12 | 2006-12-13 | 百度在线网络技术（北京）有限公司 | Advertisement display system and method used for Internet search engine |
| CN102067105A (en) * | 2008-03-18 | 2011-05-18 | 雅虎公司 | Personalizing sponsored search advertising layout using user behavior history |
| US20110184936A1 (en) * | 2010-01-24 | 2011-07-28 | Microsoft Corporation | Dynamic community-based cache for mobile search |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108713186A (en) * | 2015-11-23 | 2018-10-26 | 微软技术许可有限责任公司 | It is prioritized locative search term |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103136310B (en) | 2017-11-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9767182B1 (en) | Classification of search queries | |
| US8799260B2 (en) | Method and system for generating web pages for topics unassociated with a dominant URL | |
| US9798820B1 (en) | Classification of keywords | |
| CN105765573B (en) | Improvements in website traffic optimization | |
| US9015148B2 (en) | Suggesting related search queries during web browsing | |
| US11455313B2 (en) | Systems and methods for intelligent prospect identification using online resources and neural network processing to classify organizations based on published materials | |
| US8577875B2 (en) | Presenting search results ordered using user preferences | |
| US8417692B2 (en) | Generalized edit distance for queries | |
| US20060155751A1 (en) | System and method for document analysis, processing and information extraction | |
| US20110119262A1 (en) | Method and System for Grouping Chunks Extracted from A Document, Highlighting the Location of A Document Chunk Within A Document, and Ranking Hyperlinks Within A Document | |
| US20080243480A1 (en) | System and method for determining semantically related terms | |
| CN102722501B (en) | Search engine and realization method thereof | |
| CN102722498A (en) | Search engine and implementation method thereof | |
| US20090192983A1 (en) | Method and system for mining, ranking and visualizing lexically similar search queries for advertisers | |
| CN102722499B (en) | Search engine and implementation method thereof | |
| US20080243826A1 (en) | System and method for determining semantically related terms | |
| US20100125781A1 (en) | Page generation by keyword | |
| US20180096067A1 (en) | Creation and optimization of resource contents | |
| CN1996316A (en) | Search engine searching method based on web page correlation | |
| US9542491B2 (en) | Utilizing keystroke logging to determine items for presentation | |
| US20140181096A1 (en) | Entity name disambiguation | |
| CN105354344A (en) | SEO (search engine optimization) system and method | |
| US20180039643A1 (en) | Analysis and management of resources in a network | |
| CN103136310A (en) | Method for conversion of new queries based on historical queries of specific websites | |
| Xu et al. | Generating personalized web search using semantic context |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: MICROSOFT TECHNOLOGY LICENSING LLC Free format text: FORMER OWNER: MICROSOFT CORP. Effective date: 20150727 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20150727 Address after: Washington State Applicant after: Micro soft technique license Co., Ltd Address before: Washington State Applicant before: Microsoft Corp. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |