CN100446003C - Blog search and browsing system of intention driven - Google Patents
Blog search and browsing system of intention driven Download PDFInfo
- Publication number
- CN100446003C CN100446003C CNB200710036340XA CN200710036340A CN100446003C CN 100446003 C CN100446003 C CN 100446003C CN B200710036340X A CNB200710036340X A CN B200710036340XA CN 200710036340 A CN200710036340 A CN 200710036340A CN 100446003 C CN100446003 C CN 100446003C
- Authority
- CN
- China
- Prior art keywords
- blog
- search
- article
- information
- module
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A blog search of intention driving includes utilizing interactive boundary module to receive information requirement inputted by user, utilizing said boundary to realize interactive operation of user to blog essay search module and blog search module as well as single blog internal browsing module and to receive final results from three said modules for arranging blog content concerned by user mostly at front so as to raise correction ratio of blog search.
Description
Technical field
The present invention relates to a kind of system of areas of information technology, specifically is Blog Search and browing system that a kind of purpose drives.
Background technology
The appearance of Web2.0 notion makes internet new media development enter the new stage.Blog is used the development that is promoting whole internet industry as the typical case under the Web2.0 notion.CNNIC (CNNIC) shows in " the China Internet network state of development statistical report " of in July, 2006 issue: often use the user of the relevant network application of blog to occupy 23.7% in China 12,300 ten thousand Internet users, as a kind of Newly Sprouted Things, blog is in developing period at a high speed.Simultaneously, blog has become a kind of important information sources on the internet gradually.How to organize, retrieval, thus effectively utilize blog data to cause the extensive concern that comprises scientific research circle and industry member.
Find by retrieval: http://blogsearch.***.com sorts to Search Results by the correlativity of analysis search word and blog or blog articles and according to the degree of correlation or creation-time; Http:// www.technorati.com, http://www.pubsub.com, http://www.bloglines.com utilize the annexation between blog to calculate the weight of blog, can sort to Search Results according to these weights; Http:// www.bloglines.com can also arrange Search Results according to the subscribed quantity of blog.Search word in the proposition Blog Search such as GiladMishne generally all is an information, compare with the user of general internet search engine, the user of Blog Search is interested in the search of the information of relevant famous entity, for example, name, ProductName, organization name, place names etc., they also are partial to current hot news, the search of the content that political event and certain art are relevant.
Yet current Blog Search product all can not satisfy the specific demand of user when using Blog Search preferably.In fact, the main article that has two types in the blog, the article of the article of information and emotion type, and present searching products does not all utilize the characteristic of Blog content itself to arrange Search Results, the result who comes the front might not be that the user needs most, thereby make that the correlativity of Search Results and user's request is lower, make search have lower precision ratio.
Present blog services provider website, for example Windows Live Spaces just shows all articles of a blog according to the inverted order of blog articles creation-time.In fact, the viewer has different information requirements to different blogs, for example, browse the blog of the best of friends of oneself, the viewer can tend to browse the article of emotion class, browses certain domain expert's blog, and the viewer can tend to the article of browsing information type class.Like product does not utilize the characteristic of Blog content itself equally, can not satisfy the specific demand of user to information in the blog preferably.
Summary of the invention
The objective of the invention is to overcome deficiency of the prior art, the Blog Search and the browing system that provide a kind of purpose to drive.Make it have function according to user's search purpose optimization searching result's ordering, thereby improve the correlativity of Search Results and user's request, improve the precision ratio of Blog Search, have the function of showing the content of a blog according to user's the purpose of browsing simultaneously, satisfy the information requirement of the personalization of different user.
The present invention is achieved by the following technical solutions, the present invention includes: interactively interface module, blog articles belong to confidence calculations module, blog articles search module, the Blog Search module of information and emotion type, module browsed in the article of single blog inside.
Described interactively interface module, as the employed graphic user interface of user, the information requirement of accepting user's input (comprises search word, the search purpose, the blog name that need browse and browse purpose), can realize that by this interface user and blog articles search module, Blog Search module and single blog inside browses the interactive operation of module, and accept the end product of above-mentioned three kinds of modules in the interactive operation process.Make the user obtain the Blog content that conforms to the self-demand purpose fast thus.
Described blog articles belongs to the confidence calculations module of information and emotion type, the blog articles set of module input browsed in the article of accepting blog articles search module, Blog Search module and single blog inside, utilize to use information that the method for machine learning obtains and the blog articles in the emotion pattern type pair set to carry out classified calculating, the confidence value that the article that calculates is belonged to information and emotion type returns to the article of blog articles search module, Blog Search module and single blog inside and browses module.
Described blog articles search module, search word according to interactively interface module user's input, from indexed data, return relevant Search Results, the set of blog articles among the result is input to the confidence calculations module that blog articles belongs to information and emotion type, article in the set that the confidence calculations module of using this blog articles to belong to information and emotion type is returned belongs to the confidence value of information and emotion type class, the user's who obtains by interactively interface module the search purpose (to the desirability of information or emotion type article) and the degree of correlation of Search Results and search word, calculate the ranking value of Search Results, Search Results is sorted, and it is returned to interactively interface module demonstration.
Described Blog Search module, search word according to interactively interface module user's input, from indexed data, return relevant Search Results, the set of blog articles among the result is input to the confidence calculations module that blog articles belongs to information and emotion type, and the confidence value that article belongs to information and emotion type class in the set that the confidence calculations module of using this blog articles to belong to information and emotion type is returned calculates the shared ratio of information article in each blog.The usage ratio value, the user's who obtains by interactively interface module the search purpose (to the desirability of information or emotion type article) and the degree of correlation of Search Results and search word, calculate the ranking value of Search Results, Search Results is sorted, and it is returned to interactively interface module demonstration.
Module browsed in the inside article of described single blog, the title of the blog of browsing according to the needs of interactively interface module user input, from indexed data, return all articles of this blog, the article set is input to the confidence calculations module that blog articles belongs to information and emotion type, article in the set that the confidence calculations module of using this blog articles to belong to information and emotion type is returned belongs to the confidence value of information and emotion type class, the user's who obtains by interactively interface module purpose of browsing (to the desirability of information or emotion type article) and the time that blog articles is created, calculate the ranking value of article, article is sorted, and the result is returned to interactively interface module demonstration.
Described interactively interface module is to increase a draw runner that can allow the user drag arbitrarily to realize that the two-end-point that draw runner can reach is labeled as information class end points and emotion type class end points respectively on general search engine interface.The position of draw runner is corresponding with user's search purpose, tend to search the article of information class more the closer to information class end points representative of consumer, tend to search the article of emotion type class more the closer to emotion type class end points representative of consumer, come the search purpose of representative of consumer with λ, the value of λ is set to-1 to 1 scope, the emotion type search purpose that-1 expression is the strongest, the information search purpose that 1 expression is the strongest.
Described blog articles belongs to the confidence calculations module of information and emotion type, calculating the degree of confidence that one piece of article belongs to information and emotion type article is by article being carried out the classification of information and emotion type: described information article is meant: be similar to article on the news website with news character, describe particular technology article, introduce the article of general knowledge, to the article of the objective evaluation of the things that takes place in the society; Described emotion type article is meant: about the article of the diary type of individual affair, describe personal mood, the article of emotion; According to definition to information and emotion type article, use support vector machine method that the sample of a large amount of informations and emotion type article is carried out statistical study, set up information and emotion pattern type.When given one piece of new blog articles, utilize the model of building up to calculate the confidence level that it belongs to information article and emotion type article automatically, use S
IfRepresent this confidence level, its value is got the real number value of [0,1] scope.
Described blog articles search module, the Search Results ranking value of calculating blog articles according to user's search purpose adopts article to belong to the confidence value (S of information and emotion type
If) with the linear method that combines of original ranking value (relevance values), user's search purpose value λ and 1-| λ | respectively as combined two-part weights, wherein S
IfValue be transformed into-1 to 1 scope ,-1 expression belongs to the highest degree of confidence of emotion type, 1 expression belongs to the highest degree of confidence of information.
Described Blog Search module, calculate the information degree value and the linear method that combines of original ranking value (relevance values) that Blog Search sort result value adopts blog according to user's search purpose, user's search purpose value λ and 1-| λ | respectively as combined two-part weights.The information degree of a blog uses following mode to calculate: one piece of blog articles can be judged as two types: if S
IfValue be greater than or equal to 0.5, article belongs to information; If S
IfValue less than 0.5, article belongs to the emotion type.The information article accounts for the information degree S of a blog of ratio calculating of total article in the use blog
b, and this value is converted into-1 to 1 scope.
Module browsed in the article of described single blog inside, and the ranking value of calculating blog articles according to user's the purpose of browsing adopts article to belong to the confidence value (S of information and emotion type
If) with the linear method that combines of creation-time, user's search purpose value λ and 1-| λ | respectively as combined two-part weights, wherein S
IfValue be transformed into-1 to 1 scope ,-1 expression belongs to the highest degree of confidence of emotion type, 1 expression belongs to the highest degree of confidence of information, the time value that article is created is transformed into [0,1] scope, the creation-time value of the article that 1 expression is up-to-date, the creation-time value of the article that 0 expression is the oldest.
Characteristics of the present invention provide Blog Search and the browing system that a kind of user intent drives.Compared with prior art, the present invention has made full use of the characteristic of blog articles itself, and the mode of employing and user interactions is improved Blog Search result's rank and improved browsing mode to the article in the single blog.The classification of among the present invention blog articles being carried out information and emotion type has very high accuracy rate and divides full rate, reaches 93% and 92% respectively.The present invention can make with the maximally related Blog content of user's request and come the front, has improved the precision ratio of Blog Search, and the efficient browsed of the content of single blog, thereby has improved user experience and satisfaction greatly.
Description of drawings
Fig. 1 has described the frame diagram of system of the present invention and the relation between each functional module.
Embodiment
Below in conjunction with accompanying drawing embodiments of the invention are elaborated: present embodiment is being to implement under the prerequisite with the technical solution of the present invention, provided detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.
As shown in Figure 1, present embodiment comprises: interactively interface module, blog articles belong to the confidence calculations module of information and emotion type, the blog articles search module, and the Blog Search module, module browsed in the article of single blog inside.Interactively interface module, as the employed graphic user interface of user, the information requirement of accepting user's input (comprises search word, the search purpose, the blog name that need browse and browse purpose), can realize that by this interface user and blog articles search module, Blog Search module and single blog inside browses the interactive operation of module, and accept the end product of above-mentioned three kinds of modules in the interactive operation process.Interactively interface module is deployed in client.Blog articles belongs to the confidence calculations module of information and emotion type and mainly finishes the computing function that article belongs to the degree of confidence of information and emotion type, and this module is deployed in server end.The function of Search Results ordering when the blog articles search module is mainly finished blog articles searched for.The function of Search Results ordering when the Blog Search module is mainly finished blog searched for.The article of single blog inside is browsed module and is mainly finished the function that the article of single blog inside is browsed.More than three modules all be divided and be deployed in the client and server end.
Below will further describe the work of each module in the system by concrete application example.In client, interface module is at first accepted user's selection information, the title of the blog that user's information requirement purpose and search word or request are browsed, interface module is passed to disparate modules to these information, then information is passed to the blog articles search module if will carry out the blog articles search; Then information is passed to the Blog Search module if will carry out Blog Search; Browse module if will carry out the article of then information being passed to single blog inside of browsing of article in the single blog.Blog articles search module and Blog Search module pass to the server end part of module separately with the search word of input, from indexed data, obtain the Search Results relevant with search word, the confidence calculations module that blog articles belongs to information and emotion type is passed in the set of blog articles among the result, this module is calculated the confidence value that every piece of blog articles belongs to information and emotion type by using the information and the emotion type textual classification model that have trained, and result of calculation is returned to corresponding search module.The blog articles search module returns to this result the client part of module; The Blog Search module is at first calculated the information degree of single blog, then result of calculation is returned to the client part of module.Corresponding then search module calculates the ranking value of Search Results, and Search Results and ranking value thereof are returned to the interface module demonstration.When the user need adjust the ordering of Search Results according to the search purpose of self, interface module is accepted user's new search purpose value, it is passed to corresponding search module, search module recomputates the ranking value of Search Results in client, the result is resequenced, return to interface module and show.The title of the blog that the article of single blog inside is browsed module and browsed except the request of accepting interface module input, user's the purpose of browsing value, and blog articles of handling in the middle of it set is single blog inside, and all the other working methods are identical with the blog articles search module.
Key of the invention process comprises: the foundation of interface module, blog articles belongs to the foundation of information and emotion type textual classification model in the confidence calculations module of information and emotion type, the calculating of Search Results ranking value in the blog articles search module, the calculating of Search Results ranking value in the Blog Search module, the calculating of article ranking value in the module browsed in the inside article of single blog.Wherein the foundation of information and emotion type textual classification model is basis of the invention process.
1. the foundation at interface:
This interface is to increase the draw runner that can allow the user drag arbitrarily on general search engine interface.The two-end-point that draw runner can reach is labeled as information class end points and emotion type class end points respectively.The position of draw runner is corresponding with user's search purpose, tend to search the article of information class more the closer to information class end points representative of consumer, tend to search the article of emotion type class more the closer to emotion type class end points representative of consumer, come the search purpose of representative of consumer with λ, the value of λ is set at-1 to 1 scope, the emotion type search purpose that-1 expression is the strongest, the information search purpose that 1 expression is the strongest.This draw runner can use the Javascript technology to realize in webpage.Rearrangement for Search Results shows, can use the AJAX technology to realize.
2. the foundation of the information of blog articles and emotion type textual classification model
According to description, select a large amount of information articles and emotion type article as training sample to information and emotion type article.On training sample set, use information gain (Information Gain) method to carry out feature selecting, the speech of selecting some is as feature.Use support vector machine (Support Vector Machine) as text classification algorithm training classifier, acquired information type and emotion type disaggregated model.Disaggregated model will be exported classification results to one piece of article in the mode of probable value, describe the degree of confidence that this piece article belongs to information and emotion type class with this probable value, this value S
IfRepresent, its value is got the real number value of [0,1] scope, and 0 expression belongs to the highest degree of confidence of emotion type, and 1 expression belongs to the highest degree of confidence of information.The classification that the disaggregated model that obtains in the present embodiment carries out information and emotion type to blog articles has very high accuracy rate and divides full rate, reaches 93% and 92% respectively.
3. the calculating of Search Results ranking value in the blog articles search module
Use Lucene that blog and article thereof are set up index data.Obtain the search word of user's input, utilize Lucene to return the Search Results relevant with search word.The Search Results ranking value of calculating blog articles according to user's search purpose adopts article to belong to the confidence value (S of information and emotion type
If) with the linear method that combines of original ranking value (relevance values), user's search purpose value λ and 1-| λ | respectively as combined two-part weights, wherein S
IfValue be transformed into-1 to 1 scope ,-1 expression belongs to the highest degree of confidence of emotion type, 1 expression belongs to the highest degree of confidence of information.
4. the calculating of Search Results ranking value in the Blog Search module
Obtain the search word of user's input, utilize Lucene to return the Search Results relevant with search word.Calculate the information degree value and the linear method that combines of original ranking value (relevance values) that Blog Search sort result value adopts blog according to user's search purpose, user's search purpose value λ and 1-| λ | respectively as combined two-part weights.The information degree of a blog uses following mode to calculate: one piece of blog articles can be judged as two types: if S
IfValue be greater than or equal to 0.5, article belongs to information; If S
IfValue less than 0.5, article belongs to the emotion type.The information article accounts for the information degree S of a blog of ratio calculating of total article in the use blog
b, and this value is converted into-1 to 1 scope.
5. the calculating of article ranking value in the module browsed in the inside article of single blog
Obtain the query word of user input, all articles of the blog of utilizing Lucene to return to browse.The ranking value of calculating blog articles according to user's the purpose of browsing adopts article to belong to the confidence value (S of information and emotion type
If) with the linear method that combines of creation-time, user's search purpose value λ and 1-| λ | respectively as combined two-part weights, wherein S
IfValue be transformed into-1 to 1 scope ,-1 expression belongs to the highest degree of confidence of emotion type, 1 expression belongs to the highest degree of confidence of information, the time value that article is created is transformed into [0,1] scope, the creation-time value of the article that 1 expression is up-to-date, the creation-time value of the article that 0 expression is the oldest.
The invention process can make with the maximally related Blog content of user's request and come the front, has improved the precision ratio of Blog Search, and the efficient browsed of the content of single blog, has improved user experience and satisfaction.
Claims (7)
1, a kind of Blog Search and browing system of purpose driving, it is characterized in that, comprise: interactively interface module, blog articles belong to confidence calculations module, blog articles search module, the Blog Search module of information and emotion type, module browsed in the article of single blog inside
Described interactively interface module, accept the information requirement of user's input, browse the interactive operation of module by the article of this interface realization user and blog articles search module, Blog Search module and single blog inside, and accept the end product of above-mentioned three kinds of modules in the interactive operation process, make the user obtain the Blog content that conforms to the self-demand purpose thus;
Described blog articles belongs to the confidence calculations module of information and emotion type, the blog articles set of module input browsed in the article of accepting blog articles search module, Blog Search module and single blog inside, blog articles in the pair set is carried out classified calculating, and the confidence value that the article that calculates is belonged to information and emotion type returns to the article of blog articles search module, Blog Search module and single blog inside and browses module;
Described blog articles search module, search word according to interactively interface module user's input, from indexed data, return relevant Search Results, the set of blog articles among the result is input to the confidence calculations module that blog articles belongs to information and emotion type, article in the set that the confidence calculations module of using blog articles to belong to information and emotion type is returned belongs to the confidence value of information and emotion type class, the user's who obtains by interactively interface module the search purpose and the degree of correlation of Search Results and search word, calculate the ranking value of Search Results, Search Results is sorted, and it is returned to interactively interface module demonstration;
Described Blog Search module, search word according to interactively interface module user's input, from indexed data, return relevant Search Results, the set of blog articles among the result is input to the confidence calculations module that blog articles belongs to information and emotion type, the confidence value that article belongs to information and emotion type class in the set that the confidence calculations module of using blog articles to belong to information and emotion type is returned calculates the shared ratio of information article in each blog, the usage ratio value, the user's who obtains by interactively interface module the search purpose and the degree of correlation of Search Results and search word, calculate the ranking value of Search Results, Search Results is sorted, and it is returned to interactively interface module demonstration;
Module browsed in the inside article of described single blog, the title of the blog of browsing according to the needs of interactively interface module user input, from indexed data, return all articles of this blog, the article set is input to the confidence calculations module that blog articles belongs to information and emotion type, article in the set that the confidence calculations module of using this blog articles to belong to information and emotion type is returned belongs to the confidence value of information and emotion type class, the user's who obtains by interactively interface module purpose of browsing and the time that blog articles is created, calculate the ranking value of article, article is sorted, and the result is returned to interactively interface module demonstration.
2, Blog Search and browing system that purpose according to claim 1 drives, it is characterized in that, described interactively interface module, be on existing search engine interface, to increase a draw runner that can allow the user drag arbitrarily to realize, the two-end-point that draw runner can reach is labeled as information class end points and emotion type class end points respectively, the position of draw runner is corresponding with user's search purpose, tend to search the article of information class more the closer to information class end points representative of consumer, tend to search the article of emotion type class more the closer to emotion type class end points representative of consumer, come the search purpose of representative of consumer with λ, the value of λ is set to-1 to 1 scope, the emotion type search purpose that-1 expression is the strongest, the information search purpose that 1 expression is the strongest.
3, the Blog Search and the browing system of purpose driving according to claim 1, it is characterized in that, described blog articles belongs to the confidence calculations module of information and emotion type, and calculating the degree of confidence that one piece of article belongs to information and emotion type article is by article being carried out the classification of information and emotion type:
Described information article is meant: the article on the news website with news character, describe particular technology article, introduce the article of general knowledge, to the article of the objective evaluation of the things that takes place in the society;
Described emotion type article is meant: about the article of the diary type of individual affair, describe personal mood, the article of emotion;
According to definition to information and emotion type article, use support vector machine method that the sample of a large amount of informations and emotion type article is carried out statistical study, set up information and emotion pattern type; When given one piece of new blog articles, utilize the model of building up to calculate the degree of confidence that it belongs to information article and emotion type article automatically, use S
IfRepresent this degree of confidence, its value is got the real number value of [0,1] scope.
4, the Blog Search and the browing system of purpose driving according to claim 1, it is characterized in that, described blog articles search module, the Search Results ranking value of calculating blog articles according to user's search purpose adopts article to belong to the confidence value S of information and emotion type
IfWith the linear method that combines of original ranking value, user's search purpose value λ and 1-| λ | respectively as combined two-part weights, wherein S
IfBroken-1 to 1 the scope of being transformed into of value ,-1 expression belongs to the highest degree of confidence of emotion type, 1 expression belongs to the highest degree of confidence of information.
5, the Blog Search and the browing system of purpose driving according to claim 1, it is characterized in that, described Blog Search module, calculate the information degree value and the linear method that combines of original ranking value that Blog Search sort result value adopts blog according to user's search purpose, user's search purpose value λ and 1-| λ | respectively as combined two-part weights, the information degree of a blog uses following mode to calculate: one piece of blog articles, if article belongs to the degree of confidence S of information and emotion type
IfValue be greater than or equal to 0.5, article belongs to information; If article belongs to the degree of confidence S of information and emotion type
IfValue less than 0.5, article belongs to the emotion type; The information article accounts for the information degree S of a blog of ratio calculating of total article in the use blog
b, and this value is converted into-1 to 1 scope.
6, the Blog Search and the browing system of purpose driving according to claim 1, it is characterized in that, module browsed in the article of described single blog inside, and the ranking value of calculating blog articles according to user's the purpose of browsing adopts article to belong to the confidence value S of information and emotion type
IfWith the linear method that combines of creation-time, user's search purpose value λ and 1-| λ | respectively as combined two-part weights, wherein S
IfValue be transformed into-1 to 1 scope ,-1 expression belongs to the highest degree of confidence of emotion type, 1 expression belongs to the highest degree of confidence of information, the time value that article is created is transformed into [0,1] scope, the creation-time value of the article that 1 expression is up-to-date, the creation-time value of the article that 0 expression is the oldest.
7, the Blog Search and the browing system of purpose driving according to claim 1 is characterized in that described interactively interface module is deployed in client, and the confidence calculations module that blog articles belongs to information and emotion type is deployed in server end; The article of blog articles search module, Blog Search module, single blog inside is browsed module and all is divided and is deployed in the client and server end.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200710036340XA CN100446003C (en) | 2007-01-11 | 2007-01-11 | Blog search and browsing system of intention driven |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200710036340XA CN100446003C (en) | 2007-01-11 | 2007-01-11 | Blog search and browsing system of intention driven |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101000632A CN101000632A (en) | 2007-07-18 |
| CN100446003C true CN100446003C (en) | 2008-12-24 |
Family
ID=38692600
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200710036340XA Expired - Fee Related CN100446003C (en) | 2007-01-11 | 2007-01-11 | Blog search and browsing system of intention driven |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100446003C (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937462B (en) * | 2010-09-03 | 2016-08-24 | 中国科学院声学研究所 | Literature review automatic searching method and system |
| CN102413065B (en) * | 2010-09-19 | 2014-05-14 | 腾讯科技（深圳）有限公司 | Twitter client, twitter message aggregation method and instant messaging client |
| CN103577049B (en) * | 2012-07-24 | 2019-04-12 | 百度在线网络技术（北京）有限公司 | A kind of method, apparatus and equipment for suggesting object for providing downloading |
| CN103150367B (en) * | 2013-03-07 | 2016-01-20 | 宁波成电泰克电子信息技术发展有限公司 | A kind of Sentiment orientation analytical approach of Chinese microblogging |
| CN103218443A (en) * | 2013-04-22 | 2013-07-24 | 中山大学 | Blogging webpage retrieval system and retrieval method |
| CN105488037A (en) * | 2014-10-13 | 2016-04-13 | 萧瑞祥 | Article subjective opinion management method and related device, and computer program product |
| US10997226B2 (en) * | 2015-05-21 | 2021-05-04 | Microsoft Technology Licensing, Llc | Crafting a response based on sentiment identification |
| CN111326222A (en) * | 2020-01-21 | 2020-06-23 | 北京华峰创业科技有限公司 | Emotion analysis method, device and system based on user diary |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1535433A (en) * | 2001-07-04 | 2004-10-06 | 库吉萨姆媒介公司 | Category based, extensible and interactive system for document retrieval |
| CN1763749A (en) * | 2004-10-20 | 2006-04-26 | 微软公司 | Analyzing operational data and another data of a search system |
-
2007
- 2007-01-11 CN CNB200710036340XA patent/CN100446003C/en not_active Expired - Fee Related
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1535433A (en) * | 2001-07-04 | 2004-10-06 | 库吉萨姆媒介公司 | Category based, extensible and interactive system for document retrieval |
| CN1763749A (en) * | 2004-10-20 | 2006-04-26 | 微软公司 | Analyzing operational data and another data of a search system |
Non-Patent Citations (6)
| Title |
|---|
| 使用分类和聚类提高搜索引擎的可用性. 刘建国，黄厚宽.铁路计算机应用，第15卷第3期. 2006 |
| 使用分类和聚类提高搜索引擎的可用性. 刘建国，黄厚宽.铁路计算机应用，第15卷第3期. 2006 * |
| 基于文本分类的网页排序算法. 李信利，马军，吕月娥，张德伟.微计算机信息，第21卷第11-3期. 2005 |
| 基于文本分类的网页排序算法. 李信利，马军，吕月娥，张德伟.微计算机信息，第21卷第11-3期. 2005 * |
| 文本自动分类在搜索引擎中的应用研究. 马慧芳，郭治成.情报杂志，第12期. 2005 |
| 文本自动分类在搜索引擎中的应用研究. 马慧芳，郭治成.情报杂志，第12期. 2005 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101000632A (en) | 2007-07-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12001490B2 (en) | Systems for and methods of finding relevant documents by analyzing tags | |
| CN100446003C (en) | Blog search and browsing system of intention driven | |
| CN103064945B (en) | Based on the Situational searching method of body | |
| CN100504866C (en) | Integrative searching result sequencing system and method | |
| CN108763321B (en) | Related entity recommendation method based on large-scale related entity network | |
| JP4637969B1 (en) | Properly understand the intent of web pages and user preferences, and recommend the best information in real time | |
| CN107633430A (en) | A kind of Method of Commodity Recommendation based on community of colony | |
| CN104484431B (en) | A kind of multi-source Personalize News webpage recommending method based on domain body | |
| CN102411754A (en) | Personalized recommendation method based on commodity property entropy | |
| CN102236646A (en) | Personalized item-level vertical pagerank algorithm iRank | |
| CN106339502A (en) | Modeling recommendation method based on user behavior data fragmentation cluster | |
| CN104933239A (en) | Hybrid model based personalized position information recommendation system and realization method therefor | |
| CN101329674A (en) | System and method for providing personalized searching | |
| CN104216881A (en) | Method and device for recommending individual labels | |
| CN101853308A (en) | Method and application terminal for personalized meta-search | |
| CN110362740B (en) | Water conservancy portal information hybrid recommendation method | |
| CN101283353A (en) | Systems for and methods of finding relevant documents by analyzing tags | |
| CN109165367B (en) | News recommendation method based on RSS subscription | |
| CN102236691A (en) | Precision guided searching tool system | |
| CN102063453A (en) | Method and device for searching based on demands of user | |
| CN104063476A (en) | Social network-based content recommending method and system | |
| CN105787068A (en) | Academic recommendation method and system based on citation network and user proficiency analysis | |
| CN103294692A (en) | Information recommendation method and system | |
| CN107180078A (en) | A kind of method for vertical search based on user profile learning | |
| CN103425767B (en) | A kind of determination method and system pointing out data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081224 Termination date: 20120111 |