CN105159932B - A kind of data retrieval engine and ordering system and method - Google Patents
A kind of data retrieval engine and ordering system and method Download PDFInfo
- Publication number
- CN105159932B CN105159932B CN201510478159.9A CN201510478159A CN105159932B CN 105159932 B CN105159932 B CN 105159932B CN 201510478159 A CN201510478159 A CN 201510478159A CN 105159932 B CN105159932 B CN 105159932B
- Authority
- CN
- China
- Prior art keywords
- user
- document
- factor
- correlation
- degree
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a kind of data retrieval engine and ordering system and methods, including user management module, are used for managing user information;Database for classified storage document and responds user's request;Relatedness computation module, calculating and sequence for search result;The relatedness computation module includes user behavior statistic submodule, to document review and applies behavior for counting user preference keyword, user;Sort relatedness computation submodule, for calculating the sequence degree of correlation and arranging search result by the sequence degree of correlation.The analytical calculation of data acquisition system further progress inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC that the present invention is matched to term, the data acquisition system retrieved is set to meet user preference, user to document review and using behavior, the degree of correlation for improving search result and user demand has saved the user query time.
Description
Technical field
The present invention relates to a kind of search engine, in particular to a kind of data retrieval engine and ordering system and method.
Background technique
It is most of to be retrieved using keyword input mode during existing search engine carries out data retrieval,
Search result is ranked up according to Keywords matching degree, and it is past that different users inputs the search result that identical keyword obtains
Toward be it is the same, users ' individualized requirement and data qualitative factor are not accounted for the sortord of search result.From a large amount of
The information that oneself needs is found in return information, can waste user's long time and energy.
Summary of the invention
Present invention is primarily aimed at solving the above problems and insufficient, provides and a kind of data are improved based on user behavior information
The degree of correlation improves a kind of data retrieval engine and ordering system of retrieval accuracy.
Another main purpose of the invention is to provide a kind of data retrieval and sort method.
To achieve the above object, the technical scheme is that
A kind of data retrieval engine and ordering system, comprising:
User management module is used for managing user information;
Database for classified storage document and responds user's request;
Relatedness computation module, calculating and sequence for search result;
The relatedness computation module includes user behavior statistic submodule, is used for counting user preference keyword, user
To document review and apply behavior;
Sort relatedness computation submodule, for calculating the sequence degree of correlation and arranging search result by the sequence degree of correlation
Column.
Further, a kind of data retrieval according to claim 1 and ordering system, it is characterised in that: the sequence
The degree of correlation is determined by inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC.
Further, the inquiry degree of correlation RC is calculated using TF-IDF method,
Wherein, i is term;
TFi (d) is the frequency that term i occurs in document d;
N is the number of all documents;
DF is the number of documents containing term i.
Further, the demand degree of correlation DC is by preference key Word similarity KeySim and Behavior preference classification similarity
ClassSim is added to obtain;
The document subject matter vector DSV and user preference that the preference key Word similarity is formed by each document index are crucial
The user preference vector UPV that vocabulary is formed carries out cosine similarity calculating,
DSV(a1,w1;a2,w2;...;am,wm)
UPV(b1,w1;b2,w2;...;bn,wn)
Behavior preference classification similarity by user to document application behavior decision,
Wherein, df (ta)、df(tb)、df(tc)、df(td) it is that classification belonging to document is browsed by user, downloaded, collecting, pushing away
Recommend number.
Further, the quality of data degree of correlation QC by user to document application factor, user to document review quantity because
Son, user are added to obtain to the document scores factor,
Wherein, FDiIt is respectively represented in the number and whole documents that the document is downloaded with FD and is downloaded most documents
Download time;FLiThe browsing that most documents is browsed in the number and whole documents that the document is browsed is respectively represented with FL
Number;FFiThe collection number that most documents is collected in the number and whole documents that the document is collected is respectively represented with FF;
FRiThe recommendation number that most documents is recommended in the recommended number of the document and whole documents is respectively represented with FR;CMiWith
CM respectively represents the number that the document is commented on and is commented on the comment number of most documents in whole documents;Score is represented
The document scoring factor.
Further, the document scores factor S core scores ability US and user to the scoring e of document by userjWeighting
It is calculated,
Wherein, j is scoring number;
The user scores ability US by static factor USS, dynamic factor UDS and professional domain factor M MS weighted calculation
It is added and obtains afterwards,
Further, the static state factor USS is calculated by the conversion value SS of the age of user, title for technical personnel, educational background,
Further, the dynamic factor UDS is determined by the active degree AD of user's monthly login system number,
Wherein, friFor monthly user's login times.
Further, if field described in the affiliated professional domain of user and the document of user's scoring is same domain, have special
Industry field factor M MS, i.e. γ are 0.1;
If field described in the affiliated professional domain of user and the document of user's scoring is different field, do not have profession neck
Domain factor M MS, i.e. γ are 0.
Another technical solution of the invention is:
A kind of data retrieval and sort method, include the following steps:
Step 1, user input term i;
Step 2 matches term i with document in database, obtains matched data set;
Step 3, according to inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC, to matched data set
In document be ranked up the degree of correlation RankC calculating,
RankC=log (RC+QC+DC) (9);
Search result is ranked up by step 4 according to sequence degree of correlation RankC.
To sum up content, a kind of data retrieval engine and ordering system of the present invention and method, are matched to term
Data acquisition system further progress inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC analytical calculation, make
The data acquisition system retrieved meets user preference, user to document review and using behavior, and improving search result and user needs
The degree of correlation asked has saved the user query time.
Detailed description of the invention
Fig. 1 is the flow chart of the method for the present invention;
Fig. 2 is the frame diagram of present system;
Fig. 3 is user's scoring ability US schematic diagram of the present invention.
Specific embodiment
Present invention is further described in detail with specific embodiment with reference to the accompanying drawing:
As shown in Fig. 2, a kind of data retrieval engine and ordering system, including user management module, database and the degree of correlation
Computing module.
User management module is used for managing user information, essential record age of user, title for technical personnel, educational background, affiliated technology
The data such as field, user preference antistop list.
Database is mainly used for storing document data, and carries out Classification Management to document, and response user is directed to all kinds of documents
Application request.
Relatedness computation module is mainly used for realizing that the calculating and sequence of search result, including user behavior count submodule
Block and sequence relatedness computation submodule.
User behavior statistic submodule, be mainly used for record user to document application behavior record (browsing, downloading, collect,
Recommend), user to document review and scoring record, user's login times record etc..
Sort relatedness computation submodule, be mainly used for calculate sequence the degree of correlation and by sequence the degree of correlation by search result into
Row arrangement.
Sequence degree of correlation RankC determined by inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC, formula
It is as follows:
RankC=log (RC+QC+DC) (9)
After user inputs term i, system matches term i with document in database, obtains matched data collection
It closes, inquiry degree of correlation RC is calculated using TF-IDF method, and formula is as follows:
TFi (d) is the frequency that term i occurs in document d, and N is the number of all documents, and DF is to contain term i
Number of documents.The it is proposed of the function is based on such a hypothesis: the word significant to difference document should be those
The frequency of occurrences is sufficiently high in document, but the frequency of occurrences word few enough in other documents of entire collection of document.
Demand degree of correlation DC is added by preference key Word similarity KeySim with Behavior preference classification similarity ClassSim
It obtains.
The document subject matter vector DSV and user preference that preference key Word similarity KeySim is formed by each document index are closed
The user preference vector UPV that keyword table is formed carries out cosine similarity calculating.User preference antistop list be derived from data retrieval and
The interest tags word of the user's self-defining saved in ordering system user management module, formula are as follows:
DSV(a1,w1;a2,w2;...;am,wm)
UPV(b1,w1;b2,w2;...;bn,wn)
Behavior preference classify similarity ClassSim by user to document application behavior decision, according to user to document application
Behavior, including browsing, downloading, collection, recommendation calculate Behavior preference classification similarity ClassSim.User goes to document application
To record the user behavior statistic submodule being derived from the relatedness computation module of system, formula is as follows:
Wherein, df (ta)、df(tb)、df(tc)、df(td) it is that classification belonging to document is browsed by user, downloaded, collecting, pushing away
Recommend number.
Quality of data degree of correlation QC by user to document application factor, user to document review Quantitative factor, user to text
The shelves scoring factor is added to obtain, and formula is as follows:
Wherein, FDiIt is respectively represented in the number and whole documents that the document is downloaded with FD and is downloaded most documents
Download time;FLiThe browsing that most documents is browsed in the number and whole documents that the document is browsed is respectively represented with FL
Number;FFiThe collection number that most documents is collected in the number and whole documents that the document is collected is respectively represented with FF;
FRiThe recommendation number that most documents is recommended in the recommended number of the document and whole documents is respectively represented with FR;CMiWith
CM respectively represents the number that the document is commented on and is commented on the comment number of most documents in whole documents;Score is represented
The document scoring factor.
By above-mentioned definition it is found thatSimilarly,
User merges processing to downloading, browsing behavior to document application factor, is assigned to the arithmetic average of the two
One weight α merges processing to collection, recommendation behavior, is assigned to weight 1- α to the arithmetic average of the two.Ordinary meaning
On, 0.1 α≤0.4 ≦.That is, the influence of the downloading and browsing behavior of user for the quality of data will be lower than collection and recommendation behavior
Influence to the quality of data.User is derived from data retrieval and ordering system relatedness computation module document application behavior record
User behavior statistic submodule.
By above-mentioned definition it is found that
User is individually assigned to weight β to comment behavior to document review Quantitative factor.On ordinary meaning, 1 β≤2 ≦.User
The user behavior statistic submodule being derived from data retrieval and ordering system relatedness computation module is recorded to document review quantity.
Document scores factor S core scores ability US and user to the scoring e of document by userjWeighted calculation obtains, public
Formula is as follows:
Wherein, j is scoring number, 1≤ej≤ 5, user scores ability US as weight, and weight is higher, and user is to data
It evaluates more credible.
User's scoring ability US will affect its confidence level to data scoring, analysis core associated with user's scoring ability
Heart factor, such as age, title for technical personnel, professional domain calculate user's scoring ability US.User scores ability US by the static factor
It is added to obtain after USS, dynamic factor UDS and professional domain factor M MS weighted calculation, formula is as follows:
η、With the operator that γ is for adjusting user's evaluation capacity calculation result.In the present embodiment, 0.6 η≤0.8 ≦,γ is 0 or 0.1.
Static factor USS is calculated by the conversion value SS of the age of user, title for technical personnel, educational background, and formula is as follows:
Wherein conversion value SS is between [0,1].
Following table is static factor USS conversion value SS schematic table:
Dynamic factor UDS refers to user's evaluation ability dynamic score, and dynamic factor refers to that user can constantly occur at any time
The factor of variation determines that formula is as follows by the active degree AD of user's monthly login system number:
It is 1, fr that active degree is defined when initialiFor monthly user's login times.User's active degree and scoring ability are just
Correlation, because any active ues can improve scoring ability during application data.
Professional domain factor M MS refers to user's professional domain matching degree score, if the affiliated professional domain of user and user
When field described in the document of scoring is same domain, then having professional domain factor M MS, i.e. γ is 0.1;
If field described in the affiliated professional domain of user and the document of user's scoring is different field, do not have profession neck
Domain factor M MS, i.e. γ are 0.
A kind of data retrieval and sort method, include the following steps:
Step 1, user input term i.
Step 2 matches term i with document in database, obtains matched data set.
Step 3, according to the document index and term matching degree in matched data set, calculated and looked into using formula (1)
Ask degree of correlation RC;
User preference keyword matches the document index in matched data set, according to document index and user
Preference Keywords matching degree calculates demand degree of correlation DC using formula (2), (3);
Automatic Evaluation is carried out to the document data quality in matched data set based on user behavior information, uses formula
(4), (5), (6), (7), (8) calculate quality of data degree of correlation QC;
According to inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC, using formula (9) to matched data
Document in set is ranked up degree of correlation RankC calculating.
Search result is ranked up by step 4 according to sequence degree of correlation RankC, and sequence follows descending sort, and sequence is related
The high ranking of RankC numerical value is spent preceding, and search result is returned into user.
As described above, the plan content in conjunction with given by attached drawing, can derive similar technical solution.But it is all not take off
Content from technical solution of the present invention, according to the technical essence of the invention it is to the above embodiments it is any it is simple modification, etc.
With variation and modification, all of which are still within the scope of the technical scheme of the invention.
Claims (3)
1. a kind of data retrieval engine and ordering system characterized by comprising
User management module is used for managing user information, records age of user, title for technical personnel, educational background, technical field, use
Family preference antistop list data;
Database for classified storage document and responds user's request;
Relatedness computation module, calculating and sequence for search result;
The relatedness computation module includes user behavior statistic submodule and sequence relatedness computation submodule;
The user behavior statistic submodule, for counting user preference keyword, user to document application behavior record, user
To document review and scoring record, user's login times record, the document application behavior record includes browsing time, downloading time
Number, recommends number at collection number;
The sequence relatedness computation submodule, for calculating the sequence degree of correlation and arranging search result by the sequence degree of correlation
Column;
The sequence degree of correlation RankC determines by inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC, formula
It is as follows:
RankC=log (RC+QC+DC) (9)
The inquiry degree of correlation RC is calculated using TF-IDF method;
The demand degree of correlation DC is added by preference key Word similarity KeySim with Behavior preference classification similarity ClassSim
It obtains;
The quality of data degree of correlation QC by user to document application factor, user to document review Quantitative factor, user to text
The shelves scoring factor is added to obtain;
The document subject matter vector DSV and user preference antistop list that the preference key Word similarity is formed by each document index
The user preference vector UPV of formation carries out cosine similarity calculating, and user preference antistop list is derived from data retrieval and sequence system
The interest tags word of the user's self-defining saved in system user management module,
DSV(a1,w1;a2,w2;...;am,wm)
UPV(b1,w1;b2,w2;...;bn,wn)
For the Behavior preference classification similarity by user to document application behavior decision, user is derived from document application behavior record
User behavior statistic submodule in the relatedness computation module of system,
Wherein, df (ta)、df(tb)、df(tc)、df(td) it is the affiliated classification of document by user's browsing, downloading, collection, recommendation time
Number;
The quality of data degree of correlation QC by user to document application factor, user to document review Quantitative factor, user to text
The shelves scoring factor is added to obtain,
Wherein, FDiThe downloading time that most documents is downloaded in the number and whole documents that the document is downloaded is respectively represented with FD
Number;FLiThe browsing time that most documents is browsed in the number and whole documents that the document is browsed is respectively represented with FL;FFi
The collection number that most documents is collected in the number and whole documents that the document is collected is respectively represented with FF;FRiAnd FR
Respectively represent the recommendation number that most documents is recommended in the recommended number of the document and whole documents;CMiDistinguish with CM
Represent the comment number that most documents is commented in the number and whole documents that the document is commented on;Score represents the document
Score the factor;
User merges processing to downloading, browsing behavior to document application factor, is assigned to one to the arithmetic average of the two
Weight α merges processing to collection, recommendation behavior, is assigned to weight 1- α to the arithmetic average of the two;
User is individually assigned to weight β to comment behavior to document review Quantitative factor;
The document scores factor S core scores ability US and user to the scoring e of document by userjWeighted calculation obtains,
Wherein, j is scoring number;
The user scores ability US by phase after static factor USS, dynamic factor UDS and professional domain factor M MS weighted calculation
Add to obtain,
The static state factor USS is calculated by the conversion value SS of the age of user, title for technical personnel, educational background,
The dynamic factor UDS determines by the active degree AD of user's monthly login system number,
Wherein, friFor monthly user's login times;
If field described in the affiliated professional domain of user and the document of user's scoring is same domain, there is the professional domain factor
MMS, i.e. γ are 0.1;
If the affiliated professional domain of user and user scoring document described in field be different field, do not have professional domain because
Sub- MMS, i.e. γ are 0.
2. a kind of data retrieval engine according to claim 1 and ordering system, it is characterised in that: the side TF-IDF
Method:
Wherein, i is term;
TFi (d) is the frequency that term i occurs in document d;
N is the number of all documents;
DF is the number of documents containing term i.
3. retrieval and the sort method of a kind of system as described in claim 1, which comprises the steps of:
Step 1, user input term i;
Step 2 matches term i with document in database, obtains matched data set;
Step 3, according to inquiry degree of correlation RC, demand degree of correlation DC and quality of data degree of correlation QC, in matched data set
Document is ranked up degree of correlation RankC calculating,
RankC=log (RC+QC+DC) (9);
The inquiry degree of correlation RC is calculated using TF-IDF method;
The demand degree of correlation DC is added by preference key Word similarity KeySim with Behavior preference classification similarity ClassSim
It obtains;
The quality of data degree of correlation QC by user to document application factor, user to document review Quantitative factor, user to text
The shelves scoring factor is added to obtain;
The document subject matter vector DSV and user preference antistop list that the preference key Word similarity is formed by each document index
The user preference vector UPV of formation carries out cosine similarity calculating, and user preference antistop list is derived from data retrieval and sequence system
The interest tags word of the user's self-defining saved in system user management module,
DSV(a1,w1;a2,w2;...;am,wm)
UPV(b1,w1;b2,w2;...;bn,wn)
For the Behavior preference classification similarity by user to document application behavior decision, user is derived from document application behavior record
User behavior statistic submodule in the relatedness computation module of system,
Wherein, df (ta)、df(tb)、df(tc)、df(td) it is classification belonging to document by user's browsing time, download time, collection
Number recommends number;
The quality of data degree of correlation QC by user to document application factor, user to document review Quantitative factor, user to text
The shelves scoring factor is added to obtain,
Wherein, FDiThe downloading time that most documents is downloaded in the number and whole documents that the document is downloaded is respectively represented with FD
Number;FLiThe browsing time that most documents is browsed in the number and whole documents that the document is browsed is respectively represented with FL;FFi
The collection number that most documents is collected in the number and whole documents that the document is collected is respectively represented with FF;FRiAnd FR
Respectively represent the recommendation number that most documents is recommended in the recommended number of the document and whole documents;CMiDistinguish with CM
Represent the comment number that most documents is commented in the number and whole documents that the document is commented on;Score represents the document
Score the factor;
User merges processing to downloading, browsing behavior to document application factor, is assigned to one to the arithmetic average of the two
Weight α merges processing to collection, recommendation behavior, is assigned to weight 1- α to the arithmetic average of the two;
User is individually assigned to weight β to comment behavior to document review Quantitative factor;
The document scores factor S core scores ability US and user to the scoring e of document by userjWeighted calculation obtains,
Wherein, j is scoring number;
The user scores ability US by phase after static factor USS, dynamic factor UDS and professional domain factor M MS weighted calculation
Add to obtain,
The static state factor USS is calculated by the conversion value SS of the age of user, title for technical personnel, educational background,
The dynamic factor UDS determines by the active degree AD of user's monthly login system number,
Wherein, friFor monthly user's login times;
If field described in the affiliated professional domain of user and the document of user's scoring is same domain, there is the professional domain factor
MMS, i.e. γ are 0.1;
If the affiliated professional domain of user and user scoring document described in field be different field, do not have professional domain because
Sub- MMS, i.e. γ are 0;
Search result is ranked up by step 4 according to sequence degree of correlation RankC.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510478159.9A CN105159932B (en) | 2015-08-07 | 2015-08-07 | A kind of data retrieval engine and ordering system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510478159.9A CN105159932B (en) | 2015-08-07 | 2015-08-07 | A kind of data retrieval engine and ordering system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105159932A CN105159932A (en) | 2015-12-16 |
CN105159932B true CN105159932B (en) | 2019-06-21 |
Family
ID=54800789
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510478159.9A Active CN105159932B (en) | 2015-08-07 | 2015-08-07 | A kind of data retrieval engine and ordering system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105159932B (en) |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105468790B (en) * | 2015-12-30 | 2019-10-29 | 北京奇艺世纪科技有限公司 | A kind of comment information search method and device |
CN107092616B (en) * | 2016-11-02 | 2020-08-28 | 北京星选科技有限公司 | Object sorting method and device |
CN106997344A (en) * | 2017-03-31 | 2017-08-01 | 成都数联铭品科技有限公司 | Keyword abstraction system |
CN107133290B (en) * | 2017-04-19 | 2019-10-29 | 中国人民解放军国防科学技术大学 | A kind of Personalized search and device |
CN107273476A (en) * | 2017-06-08 | 2017-10-20 | 广州优视网络科技有限公司 | A kind of article search method, device and server |
CN107229738B (en) * | 2017-06-18 | 2020-04-03 | 杭州电子科技大学 | Academic paper search ordering method based on document scoring model and relevancy |
CN108133016A (en) * | 2017-12-22 | 2018-06-08 | 大连景竣科技有限公司 | One kind does public document alignment system and method |
CN109144954B (en) * | 2018-09-18 | 2021-03-16 | 北京字节跳动网络技术有限公司 | Resource recommendation method and device for editing document and electronic equipment |
CN111125086B (en) * | 2018-10-31 | 2023-02-24 | 北京国双科技有限公司 | Method, device, storage medium and processor for acquiring data resources |
CN111783448B (en) * | 2020-06-23 | 2024-03-15 | 北京百度网讯科技有限公司 | Document dynamic adjustment method, device, equipment and readable storage medium |
CN112100213B (en) * | 2020-09-07 | 2022-10-21 | 中国人民解放军海军工程大学 | Ship equipment technical data searching and sorting method |
CN112632383A (en) * | 2020-12-26 | 2021-04-09 | 中国农业银行股份有限公司 | Information recommendation method and device |
CN116932859A (en) * | 2023-08-10 | 2023-10-24 | 苏州阿基米德网络科技有限公司 | Medical equipment document searching and browsing method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101169797A (en) * | 2007-11-30 | 2008-04-30 | 朱廷劭 | Search result optimization method |
CN101359331A (en) * | 2008-05-04 | 2009-02-04 | 索意互动(北京)信息技术有限公司 | Method and system for reordering search result |
CN103309900A (en) * | 2012-03-06 | 2013-09-18 | 祁勇 | Personalized multidimensional document sequencing method and system |
CN103778260A (en) * | 2014-03-03 | 2014-05-07 | 哈尔滨工业大学 | Individualized microblog information recommending system and method |
-
2015
- 2015-08-07 CN CN201510478159.9A patent/CN105159932B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101169797A (en) * | 2007-11-30 | 2008-04-30 | 朱廷劭 | Search result optimization method |
CN101359331A (en) * | 2008-05-04 | 2009-02-04 | 索意互动(北京)信息技术有限公司 | Method and system for reordering search result |
CN103309900A (en) * | 2012-03-06 | 2013-09-18 | 祁勇 | Personalized multidimensional document sequencing method and system |
CN103778260A (en) * | 2014-03-03 | 2014-05-07 | 哈尔滨工业大学 | Individualized microblog information recommending system and method |
Also Published As
Publication number | Publication date |
---|---|
CN105159932A (en) | 2015-12-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105159932B (en) | A kind of data retrieval engine and ordering system and method | |
CN105224699B (en) | News recommendation method and device | |
KR101536520B1 (en) | Method and server for extracting topic and evaluating compatibility of the extracted topic | |
Liu et al. | Integrating expert profile, reputation and link analysis for expert finding in question-answering websites | |
Zhao et al. | Topical keyphrase extraction from twitter | |
EP2211280B1 (en) | System and method for providing default hierarchical training for social indexing | |
CN101408886B (en) | Selecting tags for a document by analyzing paragraphs of the document | |
US8990241B2 (en) | System and method for recommending queries related to trending topics based on a received query | |
CN103714084B (en) | The method and apparatus of recommendation information | |
JP5391633B2 (en) | Term recommendation to define the ontology space | |
CN103559206B (en) | A kind of information recommendation method and system | |
Li et al. | A two-dimensional click model for query auto-completion | |
CN107862022B (en) | Culture resource recommendation system | |
CN106339502A (en) | Modeling recommendation method based on user behavior data fragmentation cluster | |
EP2159715A2 (en) | System and method for providing a topic-directed search | |
US20090307213A1 (en) | Suffix Tree Similarity Measure for Document Clustering | |
CN104866554B (en) | A kind of individuation search method and system based on socialization mark | |
KR20150036117A (en) | Query expansion | |
CN107895303B (en) | Personalized recommendation method based on OCEAN model | |
CN105378730A (en) | Social media content analysis and output | |
CN105512300B (en) | information filtering method and system | |
CN110019763B (en) | Text filtering method, system, equipment and computer readable storage medium | |
CN106202498A (en) | A kind of network behavior custom quantization method based on classification corpus key word word frequency record association | |
Watanabe et al. | A paper recommendation mechanism for the research support system papits | |
Ganu et al. | One size does not fit all: Multi-granularity search of web forums |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information |
Address after: Jihongtan town Chengyang District Shandong city Qingdao province Jinhong road 266111 No. 88 Applicant after: CRRC QINGDAO SIFANG CO., LTD. Address before: Jihongtan town Chengyang District Shandong city Qingdao province Jinhong road 266111 No. 88 Applicant before: CSR Qingdao Sifang Locomotive and Rolling Stock Co., Ltd. |
|
COR | Change of bibliographic data | ||
GR01 | Patent grant | ||
GR01 | Patent grant |