CN103399883A - Method and system for performing personalized recommendation according to user interest points/concerns - Google Patents
Method and system for performing personalized recommendation according to user interest points/concerns Download PDFInfo
- Publication number
- CN103399883A CN103399883A CN2013103046712A CN201310304671A CN103399883A CN 103399883 A CN103399883 A CN 103399883A CN 2013103046712 A CN2013103046712 A CN 2013103046712A CN 201310304671 A CN201310304671 A CN 201310304671A CN 103399883 A CN103399883 A CN 103399883A
- Authority
- CN
- China
- Prior art keywords
- user
- focus
- session
- personalized recommendation
- point
- 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 provides a method and a system for performing personalized recommendation according to user interest points/concerns. The method includes: (a) integrating user behavior logs of a user on multiple product lines within a scheduled time period, (b) dividing sessions according to the logs of the user within the scheduled time period, (c) performing similar requirement behavior information integration on the user behavior logs of the divided sessions and mining the user interest points/concerns, and (d) displaying personalized recommendations aiming for the user interest points/concerns through a recommendation ranking model according to the mined user interest points/concerns.
Description
Technical field
The present invention relates to network service, more particularly, relate to a kind of method and system that carries out personalized recommendation according to user interest point.
Background technology
Along with the development of electronic information technology, network has changed people's life style.For example, people can utilize the interested books of Network Capture oneself, film, music, commodity etc., so Netowrk tape has been given the life of people's efficient quicks.People have been accustomed to utilizing computing machine, mobile phone etc. to have the equipment of function of surfing the Net, and own interested webpage is learnt by browsing, amusement, shopping etc. meet self multi-faceted demand.
Generally, the user finds interested webpage by the input of the search engine at network key word.Yet by the search of this search engine, can present to the webpage that the user is very many, therefore to the user, find the content of oneself expecting to bring difficulty.
Summary of the invention
The present invention seeks to realize the behavior record on a plurality of product lines from the user, the concern demand point of digging user makes up the sparse property of single product line user behavior, from a plurality of dimensions, understand fully and portray the user, thereby search and the recommendation service of better personalization are provided to the user.
The object of the present invention is to provide a kind of method and system that carries out personalized recommendation according to user interest point.
According to an aspect of the present invention, provide a kind of method of carrying out personalized recommendation according to user interest point/focus, the method comprises: (a) integrate within a predetermined period of time the User action log of user on a plurality of product lines; (b) for user's user behaviors log within a predetermined period of time, carry out sessionizing; (c) User action log of the session of dividing being carried out to similar demand behavioural information integrates and digging user focus/point of interest; (d) according to user interest point/focus of excavating, by the recommendation order models, show the personalized recommendation for user interest point/focus.
Step (b) can comprise following sub-step: for user's user behaviors log within a predetermined period of time, carry out the piece division; The session at the set of blocks place by judging division with sorter.
Described division can be carried out based at least one in following rule: rule 1: the time, identical continuous daily record was classified as same; Rule 2: if 2 continuous daily records are classified as different pieces interval greater than certain hour; Rule 3: if the product line of daily record is specific product line continuously, it is classified as to same; Rule 4: whether identically contrast the text message that extracts in the text message that extracts in current daily record and a upper daily record, if the same be classified as same, otherwise be the beginning of a new piece.
The step of the session at the set of blocks place that judgement is divided can comprise: the set of blocks of dividing by traversal, for given current block, at first judges whether and a upper piece belongs in a session; If so, this piece is included into to the session at previous place; Otherwise utilize Backtracking Strategy, judge in all pieces in the schedule time that looks forward whether the piece that belongs to a session with current block is arranged, if having, current block is included in the session at the piece place of finding, otherwise current block is included in new session.
Step (c) can comprise following sub-step: the specific fields in the session of dividing by extraction is calculated target data; Utilize the subject classification technology to calculate the theme of each target, and the target merger that each subject classification is identical is together to form a plurality of goal sets; Each goal set is carried out to participle to set up term co-occurrence figure; By the application drawing clustering algorithm, identify user interest point/focus.
Displaying can comprise for the step of the personalized recommendation of user interest point/focus: when the user searched in search engine, the personalized recommendation that will meet user interest point/focus based on the search word of user input was presented at the front of Search Results.
Displaying can comprise for the step of the personalized recommendation of user interest point/focus: when user's browsing page, display symbol is share the personalized recommendation of family point of interest/focus in all the other blocks of the main blocks of removing the webpage of browsing.
According to a further aspect in the invention, provide a kind of system of carrying out personalized recommendation according to user interest point/focus, this system comprises: the integral data source module, be used to integrating within a predetermined period of time the User action log of user on a plurality of product lines; The sessionizing module, for carrying out sessionizing for user's user behaviors log within a predetermined period of time; Focus/point of interest excavates module, for the User action log of the session to dividing, carries out similar demand behavioural information and integrates and digging user focus/point of interest; The personalized recommendation module, show the personalized recommendation for user interest point/focus for the user interest point/focus according to excavating by the recommendation order models.
The sessionizing module can comprise: piece is divided submodule, for for user's user behaviors log within a predetermined period of time, carrying out the piece division; The sessionizing submodule, for the session at the set of blocks place by judge division with sorter.
Piece is divided submodule and can be come execution block to divide based at least one in following rule: rule 1: the time, identical continuous daily record was classified as same; Rule 2: if 2 continuous daily records are classified as different pieces interval greater than certain hour; Rule 3: if the product line of daily record is specific product line continuously, it is classified as to same; Rule 4: whether identically contrast the text message that extracts in the text message that extracts in current daily record and a upper daily record, if the same be classified as same, otherwise be the beginning of a new piece.
The sessionizing submodule can judge by following steps the session at the set of blocks place of division: the set of blocks of dividing by traversal, for given current block, at first judges whether and a upper piece belongs in a session; If so, this piece is included into to the session at previous place; Otherwise utilize Backtracking Strategy, judge in all pieces in the schedule time that looks forward whether the piece that belongs to a session with current block is arranged, if having, current block is included in the session at the piece place of finding, otherwise current block is included in new session.
Focus/point of interest excavates module and can comprise: the target data calculating sub module, for the specific fields of the session of dividing by extraction, calculate target data; Goal set forms submodule, be used to utilizing the subject classification technology, calculate the theme of each target, and the target merger that each subject classification is identical is together to form a plurality of goal sets; Term co-occurrence figure sets up submodule, for each goal set being carried out to participle to set up term co-occurrence figure; User interest point/focus recognin module, for identifying user interest point/focus by the application drawing clustering algorithm.
When the user searches in search engine, the personalized recommendation that can will meet by the search word based on user's input user interest point/focus is presented at the front of Search Results, and the personalized recommendation module is showed the personalized recommendation for user interest point/focus.
When user's browsing page, can by display symbol in all the other blocks of the main blocks removing the webpage of browsing, share the personalized recommendation of family point of interest/focus, the personalized recommendation module is showed the personalized recommendation for user interest point/focus.
Will be in ensuing description part set forth the present invention other aspect and/or advantage, some will be clearly by describing, or can learn through enforcement of the present invention.
The accompanying drawing explanation
By the description of carrying out below in conjunction with accompanying drawing, above and other purpose of the present invention and characteristics will become apparent, wherein:
Fig. 1 illustrates according to user interest point, to carry out the process flow diagram of the method for personalized recommendation according to the embodiment of the present invention;
Fig. 2 illustrates the process flow diagram that session divides;
Fig. 3 illustrates the example that session divides;
Fig. 4 illustrates User action log to the session that divides to carry out that similar demand behavioural information is integrated and the process flow diagram of the process of digging user focus/point of interest;
Fig. 5 carries out the block diagram of the system of personalized recommendation according to the embodiment of the present invention according to user interest point/focus.
Embodiment
Now, describe embodiments of the invention in detail, its example represents in the accompanying drawings, and wherein, identical label represents identical parts all the time.Below by reference to accompanying drawing, embodiment is described to explain the present invention.
Fig. 1 illustrates according to user interest point/focus, to carry out the process flow diagram of the method for personalized recommendation according to the embodiment of the present invention.
As shown in Figure 1, at step S101, integrate a plurality of data sources, namely integrate within a predetermined period of time the User action log (for example, search click user behaviors log) of user on a plurality of product lines.The product line here for example can comprise large search, mhkc, knows, library etc.Specifically, by take user identity (ID) as keyword, according to time sequencing, coming the search of organizing user on each product line to click user behaviors log.
At step S102, for user's user behaviors log within a predetermined period of time, carry out session (session) and divide.
Specifically, session (session) is a logical meaning, and it represents the intention of a user within certain period, and from user's daily record behavior, session specifically can stipulations become to have the one group of retrieval that is associated and click.For example, a user has searched for 6 inquiries of " Beijing fresh flower " " BMW " " fresh flower express delivery " " benz " " Buick " " fresh flower purchase " continuously.Wherein we can find out<" Beijing fresh flower " " fresh flower express delivery " " fresh flower purchase " > and<" BMW " " benz " " Buick " > these two groups retrievals belong to different behavior intention, are two different session.Therefore, same session can be defined as same user and for the search that meets certain single piece of information demand, click behavior within one period continuous time.
Fig. 2 illustrates the process flow diagram that session divides.
At step S201, for each user user behaviors log within a predetermined period of time, carry out piece (block) and divide.Described block divides and carries out based on following rule:
Rule 1: the time, identical continuous daily record was classified as same block;
Rule 2: if 2 continuous daily records are classified as different block interval greater than certain hour (for example, 5 minutes);
Rule 3: if the product line of daily record is to be specific product line (for example " news ", " ting ", " map ") continuously, it is classified as on same block(general knowledge and thinks if inherent reading people's continuous time news, tin song are arranged or browse map think same intention, therefore be classified as same block);
Rule 4: whether identically contrast the text message that extracts in the text message that extracts in current daily record and a upper daily record, if the same be classified as same block, otherwise be the beginning of a new block.
Should be appreciated that, sequencing according to above-mentioned rule (rule 1 is to rule 4) judges whether to belong to same block for user journal, judges first namely whether user journal meets rule 1, if do not meet rule 1, continue judgment rule 2, by that analogy.
At step S202, by the block that judges division with sorter, gather the session at place.Because sorter belongs to the prior art in machine learning, therefore at this, do not repeat.
Specifically, the block set of dividing by traversal, for given current block, at first judge whether and a upper block belongs in a session.If so, this block is included into to the session at previous block place; Otherwise utilize Backtracking Strategy, judge in all block in look forward the schedule time (for example hour) whether the block that belongs to a session with current block is arranged, if have current block be included in the session at the block place of finding, otherwise current block is included in new session.More particularly, whether Backtracking Strategy mainly after judging whether 2 block belong to same session, then belongs to same session with current block toward the block that reviews within the schedule time (for example one hour).If not reopen a new session, otherwise record current block and block before is identical session id.
The example that below provides with reference to Fig. 3 is described the session partition process in detail.
As shown in Figure 3, this user carries out " Man U " inquiry at time 20:19:14, at 20:21:38, carries out " Man U is live " inquiry, at 22:01:04, carries out " position of long small pox " inquiry, at 22:11:51, carries out " how improving employee's instability " inquiry, at 22:19:11, carries out " long small pox is what is eaten " inquiry, at 23:02:44, carries out " how stablizing Staff of Employees " inquiry.
According to above-mentioned block, divide, can obtain 6 middle block of Fig. 3, then travel through this 6 block, for current block, utilize sorter to judge whether and a upper block belongs to session, with sorter judgement block1(Man U) and block2(Man U live) belong in a session.Then consider block3 and block2, according to sorter, determine that these two block do not belong to same session, next back consider block3 and block1, due to these two block interval greater than the schedule time (namely 1 hour), therefore block3 belongs to new session.When should traverse block5 together, find that itself and block4 do not belong to same session, toward the block that reviews in hour, the time interval of discovery and block4, block3 is all within one hour, therefore block5 and block4, block3 are judged, by sorter, find that block5 and block4 be not in a session, and belong to a session together with block3, block5 is included in the session at block3 place.By that analogy, find that block6 and block4 belong to a session together.After having traveled through all block, the result on the right in output map 3, and result is divided for the session that does not consider Backtracking Strategy in the left side.
Now get back to Fig. 1, at step S103, the User action log of the session that divides is carried out to similar demand behavioural information and integrate and digging user focus/point of interest.
Below with reference to Fig. 4, step S103 is described in detail.
At step S401, the useful field in the session (session) of dividing by extraction is calculated target (goal) data.
Specifically, characteristics for the different pieces of information source, because the daily record difference of different product lines, so by with Different Strategies, extracting the target data that generates the user to calculating the useful field of point of interest, so the useful field here can refer to inquiry (Query), title (title) etc.Query Information when calculating target in the general inquiry field can twice be processed, other are all one times of processing, this is more can express user interest because we think to inquire about, and the non-Query Information fields such as header field (the title here refers to the web page title of clicking after user search) are just supplemented the useful of inquiry.
At step S402, utilize the subject classification technology to calculate the theme of each target, and the target merger that each user's subject classification is identical is together to form a plurality of goal sets.The subject classification technology is to user search queries or clicks the machine learning techniques that title is classified.Input is an inquiry or title text, is output as some classifications that this inquiry or title can be assigned to, such as " amusement personage ", and " traffic ticketing service ", " educational training " etc.This subject classification technology is the routine techniques in machine learning, therefore at this, is not repeated.
At step S403, each goal set is carried out to participle to set up term (term) co-occurrence figure.
Specifically, inquiry and the title of the user in each goal set are cut to word, according to the term co-occurrence, set up term co-occurrence figure.The corresponding term of node in figure in each term co-occurrence figure, the frequency that the corresponding term of the weight of node occurs.The co-occurrence of corresponding two terms in every limit of co-occurrence figure, the frequency of corresponding two the term co-occurrences of the weight of connection.For example, take inquiry " I like dog " as example, after cutting word, become " I ", " liking ", " dog " these words, the figure of that foundation has three nodes, word of each node, weight on node is exactly the frequency that these words occur, the weight on the limit of connection is exactly the common number of times that occur of two words, for example, I and like jointly having occurred 1 time.
At step S404, by the application drawing clustering algorithm, identify user interest point/focus.
Specifically, figure carries out Agglomerative Hierarchical Clustering to the term co-occurrence, with identify in term co-occurrence figure total what Cluster (bunch), think a user's of each bunch correspondence point of interest here.
, by next step respectively export point of interest in each cluster result: while extracting user interest point, first be chosen in the core term of the node of weight maximum in this bunch as this point of interest thereafter.Then choose connection weight and multiply by the maximum and term that is connected with core node of node weights as the first auxiliary term.Finally all are all extracted as other auxiliary terms with the term that the first auxiliary term is connected with the core term.So the term quantity of each point of interest is uncertain, if be non-isolated node, term quantity is necessarily more than or equal to 2; If isolated node, term quantity is 1.
Afterwards, get back in original Session and again check point of interest and whether can export: if point of interest at many days, occurred or appeared more than 3 times at Dan Tian, think that behavior is enough abundant, can export, otherwise not export.
In addition, each point of interest of output comprises the core term, auxiliary term etc.Each point of interest is also exported the product line in source, with assistance application quadrate part administration policy filtering strategy; Also export simultaneously the place name term, thus the region tendency of identification interest; And export its last and the date occurs and number of days always occurs, ageing etc. with assistance application side judgement point of interest.
Then, utilize the dictionary that has built, the point of interest of having exported is crossed to the dictionary matching module, with this point of interest is tagged (tag).The dictionary matching strategy is only considered core term and the first auxiliary term, at first these two terms is connected to (comprising forward and reverse), carries out dictionary matching.If the match is successful, the label of output matching; Otherwise utilize the core term to mate, if the match is successful, the label of output matching; Otherwise utilize the first auxiliary term to mate, if the match is successful, the label of output matching.
Return to now Fig. 1, at step S104, according to user interest point/focus of excavating, by the recommendation order models, show the personalized recommendation for user interest point/focus.
Above-mentioned steps S101, S102 and S103 can carry out or carry out in server in user terminal.Specifically, user terminal can be downloaded corresponding user interest point/focus from server and excavate software so that digging user point of interest/focus.According to above description, user interest point/focus excavates software can constantly upgrade user interest point/focus according to user's inquiry click Operation Log.
Above-mentioned steps S104 can carry out on server.Above-mentioned recommendation order models can be realized by existing recommendation order models, and for example, existing recommendation order models can be bidded according to businessman, according to IP address etc. condition, Search Results be sorted, and therefore at this, is not described in detail.
According to embodiments of the invention, for example, the website, article, video etc. of recommending order models can will meet based on the search word of user's input user interest point/focus when the user searches in search engine are listed in the front of Search Results.Perhaps, can be when user's browsing page, the website that demonstration is recommended according to user interest point/focus in all the other blocks of the main blocks of removing the webpage of browsing, article, video etc., for example, when user interest point/focus is stock, can in all the other blocks of the webpage that the user browses, for the user, recommend the website relevant to stock, article, video etc.
Fig. 5 carries out the block diagram of the system of personalized recommendation according to the embodiment of the present invention according to user interest point/focus.
As shown in Figure 5, system of carrying out personalized recommendation according to user interest point according to the present invention comprises: integral data source module 501, sessionizing module 502, focus/point of interest excavate module 503 and personalized recommendation module 504.
Integral data source module 501 is be used to integrating within a predetermined period of time the User action log of user on a plurality of product lines.
Focus/point of interest excavation module 503 is carried out similar demand behavioural information integration and digging user focus/point of interest for the User action log of the session to dividing.
Should be appreciated that, integral data source module 501, sessionizing module 502, focus/point of interest excavate module 503 can be by independent or integration realization in user terminal or integration realization on server.Personalized recommendation module 504 is implemented on server.
As shown in Figure 5, sessionizing module 502 comprises piece division submodule 5021 and sessionizing submodule 5022.
Wherein, piece is divided submodule 5021 for for each user user behaviors log within a predetermined period of time, carrying out piece (block) division.Described block divides and carries out based on following rule:
Rule 1: the time, identical continuous daily record was classified as same block;
Rule 2: if 2 continuous daily records are classified as different block interval greater than certain hour (for example, 5 minutes);
Rule 3: if the product line of daily record is to be specific product line (for example " news ", " ting ", " map ") continuously, it is classified as on same block(general knowledge and thinks if inherent reading people's continuous time news, tin song are arranged or browse map think same intention, therefore be classified as same block);
Rule 4: whether identically contrast the text message that extracts in the text message that extracts in current daily record and a upper daily record, if the same be classified as same block, otherwise be the beginning of a new block.
Sessionizing submodule 5022 is for the session at the block set place by judge division with sorter.
And for example shown in Figure 5, focus/point of interest excavates module 503 and comprises that target data calculating sub module 5031, goal set form submodule 5032, term co-occurrence figure sets up submodule 5033 and user interest point/focus recognin module 5034.
Target data calculating sub module 5031 is calculated target (goal) data for the useful field of the session (session) of dividing by extraction.
Goal set forms submodule 5032 be used to utilizing the subject classification technology to calculate the theme of each target, and the target merger that each user's subject classification is identical is together to form a plurality of goal sets.
Term co-occurrence figure sets up submodule 5033 for each goal set being carried out to participle to set up term co-occurrence figure.
User interest point/focus recognin module 5034 is for identifying user interest point/focus by the application drawing clustering algorithm.
By the method and system that carries out personalized recommendation according to user interest point/focus according to the present invention, by determining point of interest/focus of user, thereby to the user, recommend more to meet the product of user interest point/focus, thereby improve user's satisfaction.
Although with reference to the embodiment of the present invention, specifically shown and described the present invention, but it should be appreciated by those skilled in the art, in the situation that do not break away from the spirit and scope of the present invention that are defined by the claims, can carry out the various changes on form and details to it.
Claims (14)
1. method of carrying out personalized recommendation according to user interest point/focus, the method comprises:
(a) integrate within a predetermined period of time the User action log of user on a plurality of product lines;
(b) for user's user behaviors log within a predetermined period of time, carry out sessionizing;
(c) User action log of the session of dividing being carried out to similar demand behavioural information integrates and digging user focus/point of interest;
(d) according to user interest point/focus of excavating, by the recommendation order models, show the personalized recommendation for user interest point/focus.
2. the method for claim 1, wherein step (b) comprises following sub-step:
For user's user behaviors log within a predetermined period of time, carry out the piece division;
The session at the set of blocks place by judging division with sorter.
3. method as claimed in claim 2, wherein, described division carried out based at least one in following rule:
Rule 1: the time, identical continuous daily record was classified as same;
Rule 2: if 2 continuous daily records are classified as different pieces interval greater than certain hour;
Rule 3: if the product line of daily record is specific product line continuously, it is classified as to same;
Rule 4: whether identically contrast the text message that extracts in the text message that extracts in current daily record and a upper daily record, if the same be classified as same, otherwise be the beginning of a new piece.
4. method as claimed in claim 2, wherein, the step of the session at the set of blocks place that judgement is divided comprises:
The set of blocks of dividing by traversal, for given current block, at first judge whether and a upper piece belongs in a session;
If so, this piece is included into to the session at previous place; Otherwise utilize Backtracking Strategy, judge in all pieces in the schedule time that looks forward whether the piece that belongs to a session with current block is arranged, if having, current block is included in the session at the piece place of finding, otherwise current block is included in new session.
5. the method for claim 1, wherein step (c) comprises following sub-step:
Specific fields in the session of dividing by extraction is calculated target data;
Utilize the subject classification technology to calculate the theme of each target, and the target merger that each subject classification is identical is together to form a plurality of goal sets;
Each goal set is carried out to participle to set up term co-occurrence figure;
By the application drawing clustering algorithm, identify user interest point/focus.
6. show the method for claim 1, wherein that the step for the personalized recommendation of user interest point/focus comprises:
When the user searched in search engine, the personalized recommendation that will meet user interest point/focus based on the search word of user input was presented at the front of Search Results.
7. show the method for claim 1, wherein that the step for the personalized recommendation of user interest point/focus comprises:
When user's browsing page, display symbol is share the personalized recommendation of family point of interest/focus in all the other blocks of the main blocks of removing the webpage of browsing.
8. system of carrying out personalized recommendation according to user interest point/focus, this system comprises:
The integral data source module, be used to integrating within a predetermined period of time the User action log of user on a plurality of product lines;
The sessionizing module, for carrying out sessionizing for user's user behaviors log within a predetermined period of time;
Focus/point of interest excavates module, for the User action log of the session to dividing, carries out similar demand behavioural information and integrates and digging user focus/point of interest;
The personalized recommendation module, show the personalized recommendation for user interest point/focus for the user interest point/focus according to excavating by the recommendation order models.
9. system as claimed in claim 8, wherein, the sessionizing module comprises:
Piece is divided submodule, for for user's user behaviors log within a predetermined period of time, carrying out the piece division;
The sessionizing submodule, for the session at the set of blocks place by judge division with sorter.
10. system as claimed in claim 9, wherein, piece is divided submodule and is come execution block to divide based at least one in following rule:
Rule 1: the time, identical continuous daily record was classified as same;
Rule 2: if 2 continuous daily records are classified as different pieces interval greater than certain hour;
Rule 3: if the product line of daily record is specific product line continuously, it is classified as to same;
Rule 4: whether identically contrast the text message that extracts in the text message that extracts in current daily record and a upper daily record, if the same be classified as same, otherwise be the beginning of a new piece.
11. system as claimed in claim 9, wherein, the sessionizing submodule judges the session at the set of blocks place of division by following steps:
The set of blocks of dividing by traversal, for given current block, at first judge whether and a upper piece belongs in a session;
If so, this piece is included into to the session at previous place; Otherwise utilize Backtracking Strategy, judge in all pieces in the schedule time that looks forward whether the piece that belongs to a session with current block is arranged, if having, current block is included in the session at the piece place of finding, otherwise current block is included in new session.
12. system as claimed in claim 8, wherein, focus/point of interest excavates module and comprises:
The target data calculating sub module, calculate target data for the specific fields of the session of dividing by extraction;
Goal set forms submodule, be used to utilizing the subject classification technology, calculate the theme of each target, and the target merger that each subject classification is identical is together to form a plurality of goal sets;
Term co-occurrence figure sets up submodule, for each goal set being carried out to participle to set up term co-occurrence figure;
User interest point/focus recognin module, for identifying user interest point/focus by the application drawing clustering algorithm.
13. system as claimed in claim 8, wherein, when the user searches in search engine, the personalized recommendation that will meet user interest point/focus by the search word based on user input is presented at the front of Search Results, and the personalized recommendation module is showed the personalized recommendation for user interest point/focus.
14. system as claimed in claim 8, wherein, when user's browsing page, by display symbol in all the other blocks of the main blocks removing the webpage of browsing, share the personalized recommendation of family point of interest/focus, the personalized recommendation module is showed the personalized recommendation for user interest point/focus.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310304671.2A CN103399883B (en) | 2013-07-19 | 2013-07-19 | Method and system for performing personalized recommendation according to user interest points/concerns |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310304671.2A CN103399883B (en) | 2013-07-19 | 2013-07-19 | Method and system for performing personalized recommendation according to user interest points/concerns |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103399883A true CN103399883A (en) | 2013-11-20 |
CN103399883B CN103399883B (en) | 2017-02-08 |
Family
ID=49563513
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310304671.2A Active CN103399883B (en) | 2013-07-19 | 2013-07-19 | Method and system for performing personalized recommendation according to user interest points/concerns |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103399883B (en) |
Cited By (38)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103699630A (en) * | 2013-12-20 | 2014-04-02 | 北京奇虎科技有限公司 | Method and device for recommending information in browser |
CN103870541A (en) * | 2014-02-24 | 2014-06-18 | 微梦创科网络科技(中国)有限公司 | Social network user interest mining method and system |
CN104462248A (en) * | 2014-11-19 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing webpage content and server |
CN104750760A (en) * | 2013-12-31 | 2015-07-01 | 中国移动通信集团上海有限公司 | Application software recommending method and device |
CN104933049A (en) * | 2014-03-17 | 2015-09-23 | 华为技术有限公司 | Method and system for generating digital human |
WO2015196397A1 (en) * | 2014-06-25 | 2015-12-30 | 北京百付宝科技有限公司 | Method and device for data mining based on user's search behaviour |
CN105224807A (en) * | 2015-10-14 | 2016-01-06 | 东软集团股份有限公司 | Case auditing rule extracting method and device, case checking method and system |
CN105718545A (en) * | 2016-01-18 | 2016-06-29 | 合一网络技术(北京)有限公司 | Recommendation method and device of multimedia resources |
CN105809464A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Method and device for information delivery |
CN105868360A (en) * | 2016-03-29 | 2016-08-17 | 乐视控股(北京)有限公司 | Content recommendation method and device based on voice recognition |
CN105956204A (en) * | 2016-07-01 | 2016-09-21 | 北京奇虎科技有限公司 | Session satisfaction degree evaluation method and Session satisfaction degree evaluation device |
CN106021516A (en) * | 2016-05-24 | 2016-10-12 | 百度在线网络技术(北京)有限公司 | Search method and device |
CN106293119A (en) * | 2016-07-29 | 2017-01-04 | 百度在线网络技术(北京)有限公司 | A kind of method and apparatus carrying out information recommendation in input method |
CN106339502A (en) * | 2016-09-18 | 2017-01-18 | 电子科技大学 | Modeling recommendation method based on user behavior data fragmentation cluster |
WO2017063420A1 (en) * | 2015-10-12 | 2017-04-20 | 百度在线网络技术(北京)有限公司 | User demand determining method and apparatus |
CN107169014A (en) * | 2017-03-31 | 2017-09-15 | 百度在线网络技术(北京)有限公司 | POI recommends method, device, equipment and computer-readable recording medium |
CN107220312A (en) * | 2017-05-12 | 2017-09-29 | 华中科技大学 | A kind of point of interest based on co-occurrence figure recommends method and system |
WO2018018196A1 (en) * | 2016-07-24 | 2018-02-01 | 严映军 | Book interest determining method and determining system |
CN107679079A (en) * | 2017-08-29 | 2018-02-09 | 重庆邮电大学 | Method is recommended in personalized application based on cellphone subscriber's interest threedimensional model |
CN107688600A (en) * | 2017-07-12 | 2018-02-13 | 百度在线网络技术(北京)有限公司 | Knowledge point method for digging and device |
WO2018045937A1 (en) * | 2016-09-12 | 2018-03-15 | 阿里巴巴集团控股有限公司 | Page information processing system, and method and apparatus for generating pages and providing page information |
CN108197331A (en) * | 2015-01-21 | 2018-06-22 | 北京字节跳动科技有限公司 | The heuristic approach and device of user interest |
CN108335147A (en) * | 2018-02-28 | 2018-07-27 | 北京搜狐新媒体信息技术有限公司 | A kind of data analysing method and system based on user behavior |
CN108345698A (en) * | 2018-03-22 | 2018-07-31 | 北京百度网讯科技有限公司 | Article focus method for digging and device |
CN108427756A (en) * | 2018-03-16 | 2018-08-21 | 中国人民解放军国防科技大学 | Personalized query word completion recommendation method and device based on same-class user model |
CN109214923A (en) * | 2018-08-02 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of financing task creating method and device |
CN109657150A (en) * | 2018-12-25 | 2019-04-19 | 杭州铭智云教育科技有限公司 | A method of target literature register is pushed to target user |
CN109902706A (en) * | 2018-11-09 | 2019-06-18 | 华为技术有限公司 | Recommended method and device |
CN110022484A (en) * | 2018-01-08 | 2019-07-16 | 武汉斗鱼网络科技有限公司 | A kind of methods of exhibiting, device, server and medium |
CN110209937A (en) * | 2019-05-31 | 2019-09-06 | 三角兽(北京)科技有限公司 | Information-pushing method and device based on user behavior |
CN111949865A (en) * | 2020-08-10 | 2020-11-17 | 杭州电子科技大学 | Interest point recommendation method based on graph neural network and user long-term and short-term preference |
CN112148960A (en) * | 2019-06-27 | 2020-12-29 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for determining category of attention point |
CN112765400A (en) * | 2020-12-31 | 2021-05-07 | 上海众源网络有限公司 | Weight updating method of interest tag, content recommendation method, device and equipment |
CN113111251A (en) * | 2020-01-10 | 2021-07-13 | 阿里巴巴集团控股有限公司 | Project recommendation method, device and system |
CN113761383A (en) * | 2021-10-13 | 2021-12-07 | 中国人民大学 | Search and recommendation fusion system based on unified user behavior modeling |
CN114818643A (en) * | 2022-06-21 | 2022-07-29 | 北京必示科技有限公司 | Log template extraction method for reserving specific service information |
CN114840486A (en) * | 2022-06-28 | 2022-08-02 | 广州趣米网络科技有限公司 | User behavior data acquisition method and system and cloud platform |
CN112765400B (en) * | 2020-12-31 | 2024-04-23 | 上海众源网络有限公司 | Weight updating method, content recommending method, device and equipment for interest labels |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6668278B1 (en) * | 1996-12-20 | 2003-12-23 | Liberate Technologies | Information retrieval system using an internet multiplexer to focus user selection |
CN102982042A (en) * | 2011-09-07 | 2013-03-20 | 中国移动通信集团公司 | Personalization content recommendation method and platform and system |
-
2013
- 2013-07-19 CN CN201310304671.2A patent/CN103399883B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6668278B1 (en) * | 1996-12-20 | 2003-12-23 | Liberate Technologies | Information retrieval system using an internet multiplexer to focus user selection |
CN102982042A (en) * | 2011-09-07 | 2013-03-20 | 中国移动通信集团公司 | Personalization content recommendation method and platform and system |
Non-Patent Citations (7)
Title |
---|
彭建光: "数据挖掘技术在网站个性化推荐中的应用", 《中国优秀硕士学位论文全文数据库·信息科技辑》 * |
方元康 等: "Web日志预处理中优化的会话识别方法", 《计算机工程》 * |
方奇 等: "基于群体智慧的web访问日志会话主题识别研究", 《中文信息学报》 * |
梅娟 等: "图聚类算法的评价模型", 《计算机工程与应用》 * |
苏欣: "Web使用挖掘的个性化推荐系统研究", 《中国优秀硕士学位论文全文数据库·信息科技辑》 * |
赵文清 等: "基于词共现图的中文微博新闻话题识别", 《智能系统学报》 * |
黄浩 等: "WEB使用挖掘研究", 《计算机系统应用》 * |
Cited By (59)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103699630B (en) * | 2013-12-20 | 2017-05-03 | 北京奇虎科技有限公司 | Method and device for recommending information in browser |
CN103699630A (en) * | 2013-12-20 | 2014-04-02 | 北京奇虎科技有限公司 | Method and device for recommending information in browser |
CN104750760A (en) * | 2013-12-31 | 2015-07-01 | 中国移动通信集团上海有限公司 | Application software recommending method and device |
CN103870541A (en) * | 2014-02-24 | 2014-06-18 | 微梦创科网络科技(中国)有限公司 | Social network user interest mining method and system |
CN104933049B (en) * | 2014-03-17 | 2019-02-19 | 华为技术有限公司 | Generate the method and system of Digital Human |
CN104933049A (en) * | 2014-03-17 | 2015-09-23 | 华为技术有限公司 | Method and system for generating digital human |
US10607133B2 (en) | 2014-03-17 | 2020-03-31 | Huawei Technologies Co., Ltd. | Digital human generation method and system |
WO2015196397A1 (en) * | 2014-06-25 | 2015-12-30 | 北京百付宝科技有限公司 | Method and device for data mining based on user's search behaviour |
CN105556514A (en) * | 2014-06-25 | 2016-05-04 | 北京百付宝科技有限公司 | Method and device for data mining based on user's search behaviour |
US10896461B2 (en) | 2014-06-25 | 2021-01-19 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for data mining based on users' search behavior |
CN104462248B (en) * | 2014-11-19 | 2019-05-24 | 北京奇虎科技有限公司 | The method, apparatus and server of web page contents are provided |
CN104462248A (en) * | 2014-11-19 | 2015-03-25 | 北京奇虎科技有限公司 | Method and device for providing webpage content and server |
CN105809464A (en) * | 2014-12-31 | 2016-07-27 | 中国电信股份有限公司 | Method and device for information delivery |
CN108197331A (en) * | 2015-01-21 | 2018-06-22 | 北京字节跳动科技有限公司 | The heuristic approach and device of user interest |
CN108197331B (en) * | 2015-01-21 | 2020-02-04 | 北京字节跳动科技有限公司 | User interest exploration method and device |
WO2017063420A1 (en) * | 2015-10-12 | 2017-04-20 | 百度在线网络技术(北京)有限公司 | User demand determining method and apparatus |
JP2018514047A (en) * | 2015-10-12 | 2018-05-31 | バイドゥ オンライン ネットワーク テクノロジー (ベイジン) カンパニー リミテッド | Method and apparatus for determining user needs |
CN105224807B (en) * | 2015-10-14 | 2018-09-14 | 东软集团股份有限公司 | Case auditing rule extracting method and device, case checking method and system |
CN105224807A (en) * | 2015-10-14 | 2016-01-06 | 东软集团股份有限公司 | Case auditing rule extracting method and device, case checking method and system |
CN105718545A (en) * | 2016-01-18 | 2016-06-29 | 合一网络技术(北京)有限公司 | Recommendation method and device of multimedia resources |
CN105868360A (en) * | 2016-03-29 | 2016-08-17 | 乐视控股(北京)有限公司 | Content recommendation method and device based on voice recognition |
CN106021516A (en) * | 2016-05-24 | 2016-10-12 | 百度在线网络技术(北京)有限公司 | Search method and device |
CN105956204B (en) * | 2016-07-01 | 2019-08-02 | 北京奇虎科技有限公司 | The method and device of session Session satisfaction assessment |
CN105956204A (en) * | 2016-07-01 | 2016-09-21 | 北京奇虎科技有限公司 | Session satisfaction degree evaluation method and Session satisfaction degree evaluation device |
WO2018018196A1 (en) * | 2016-07-24 | 2018-02-01 | 严映军 | Book interest determining method and determining system |
CN106293119A (en) * | 2016-07-29 | 2017-01-04 | 百度在线网络技术(北京)有限公司 | A kind of method and apparatus carrying out information recommendation in input method |
CN107817976A (en) * | 2016-09-12 | 2018-03-20 | 阿里巴巴集团控股有限公司 | Page info processing system and page generation, page info provide method and device |
WO2018045937A1 (en) * | 2016-09-12 | 2018-03-15 | 阿里巴巴集团控股有限公司 | Page information processing system, and method and apparatus for generating pages and providing page information |
CN106339502A (en) * | 2016-09-18 | 2017-01-18 | 电子科技大学 | Modeling recommendation method based on user behavior data fragmentation cluster |
CN107169014A (en) * | 2017-03-31 | 2017-09-15 | 百度在线网络技术(北京)有限公司 | POI recommends method, device, equipment and computer-readable recording medium |
CN107169014B (en) * | 2017-03-31 | 2021-03-19 | 百度在线网络技术(北京)有限公司 | POI recommendation method, device, equipment and computer readable storage medium |
CN107220312A (en) * | 2017-05-12 | 2017-09-29 | 华中科技大学 | A kind of point of interest based on co-occurrence figure recommends method and system |
CN107220312B (en) * | 2017-05-12 | 2020-08-14 | 华中科技大学 | Interest point recommendation method and system based on co-occurrence graph |
CN107688600A (en) * | 2017-07-12 | 2018-02-13 | 百度在线网络技术(北京)有限公司 | Knowledge point method for digging and device |
CN107679079A (en) * | 2017-08-29 | 2018-02-09 | 重庆邮电大学 | Method is recommended in personalized application based on cellphone subscriber's interest threedimensional model |
CN110022484A (en) * | 2018-01-08 | 2019-07-16 | 武汉斗鱼网络科技有限公司 | A kind of methods of exhibiting, device, server and medium |
CN108335147B (en) * | 2018-02-28 | 2021-08-13 | 北京搜狐新媒体信息技术有限公司 | Data analysis method and system based on user behaviors |
CN108335147A (en) * | 2018-02-28 | 2018-07-27 | 北京搜狐新媒体信息技术有限公司 | A kind of data analysing method and system based on user behavior |
CN108427756A (en) * | 2018-03-16 | 2018-08-21 | 中国人民解放军国防科技大学 | Personalized query word completion recommendation method and device based on same-class user model |
CN108427756B (en) * | 2018-03-16 | 2021-02-12 | 中国人民解放军国防科技大学 | Personalized query word completion recommendation method and device based on same-class user model |
CN108345698B (en) * | 2018-03-22 | 2022-03-11 | 北京百度网讯科技有限公司 | Method and device for mining attention points of articles |
CN108345698A (en) * | 2018-03-22 | 2018-07-31 | 北京百度网讯科技有限公司 | Article focus method for digging and device |
CN109214923A (en) * | 2018-08-02 | 2019-01-15 | 阿里巴巴集团控股有限公司 | A kind of financing task creating method and device |
CN109902706B (en) * | 2018-11-09 | 2023-08-22 | 华为技术有限公司 | Recommendation method and device |
CN109902706A (en) * | 2018-11-09 | 2019-06-18 | 华为技术有限公司 | Recommended method and device |
CN109657150A (en) * | 2018-12-25 | 2019-04-19 | 杭州铭智云教育科技有限公司 | A method of target literature register is pushed to target user |
CN109657150B (en) * | 2018-12-25 | 2020-07-14 | 场亿租网络科技(上海)有限公司 | Method for pushing target document directory to target user |
CN110209937A (en) * | 2019-05-31 | 2019-09-06 | 三角兽(北京)科技有限公司 | Information-pushing method and device based on user behavior |
CN112148960A (en) * | 2019-06-27 | 2020-12-29 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for determining category of attention point |
CN112148960B (en) * | 2019-06-27 | 2024-03-22 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for determining category of attention point |
CN113111251A (en) * | 2020-01-10 | 2021-07-13 | 阿里巴巴集团控股有限公司 | Project recommendation method, device and system |
CN111949865A (en) * | 2020-08-10 | 2020-11-17 | 杭州电子科技大学 | Interest point recommendation method based on graph neural network and user long-term and short-term preference |
CN112765400A (en) * | 2020-12-31 | 2021-05-07 | 上海众源网络有限公司 | Weight updating method of interest tag, content recommendation method, device and equipment |
CN112765400B (en) * | 2020-12-31 | 2024-04-23 | 上海众源网络有限公司 | Weight updating method, content recommending method, device and equipment for interest labels |
CN113761383B (en) * | 2021-10-13 | 2023-06-20 | 中国人民大学 | Search and recommendation fusion system based on unified user behavior modeling |
CN113761383A (en) * | 2021-10-13 | 2021-12-07 | 中国人民大学 | Search and recommendation fusion system based on unified user behavior modeling |
CN114818643A (en) * | 2022-06-21 | 2022-07-29 | 北京必示科技有限公司 | Log template extraction method for reserving specific service information |
CN114840486B (en) * | 2022-06-28 | 2022-09-16 | 广州趣米网络科技有限公司 | User behavior data acquisition method and system and cloud platform |
CN114840486A (en) * | 2022-06-28 | 2022-08-02 | 广州趣米网络科技有限公司 | User behavior data acquisition method and system and cloud platform |
Also Published As
Publication number | Publication date |
---|---|
CN103399883B (en) | 2017-02-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103399883A (en) | Method and system for performing personalized recommendation according to user interest points/concerns | |
US11580104B2 (en) | Method, apparatus, device, and storage medium for intention recommendation | |
Vargiu et al. | Exploiting web scraping in a collaborative filtering-based approach to web advertising. | |
CN101876981B (en) | A kind of method and device building knowledge base | |
US8650198B2 (en) | Systems and methods for facilitating the gathering of open source intelligence | |
CN102760138B (en) | Classification method and device for user network behaviors and search method and device for user network behaviors | |
CN101295319B (en) | Method and device for expanding query, search engine system | |
CN102968465B (en) | Network information service platform and the search service method based on this platform thereof | |
CN104111941B (en) | The method and apparatus that information is shown | |
CN104484431B (en) | A kind of multi-source Personalize News webpage recommending method based on domain body | |
WO2013030133A1 (en) | Search and discovery system | |
CN105677780A (en) | Scalable user intent mining method and system thereof | |
CN103023714A (en) | Activeness and cluster structure analyzing system and method based on network topics | |
CN102171689A (en) | Providing posts to discussion threads in response to a search query | |
CN105022827A (en) | Field subject-oriented Web news dynamic aggregation method | |
CN103631803A (en) | Method, device and server for advertisement orientation based on input behaviors | |
CN103365904A (en) | Advertising information searching method and system | |
CN103902697A (en) | Combinatorial search method, client and server | |
CN104778208A (en) | Method and system for optimally grasping search engine SEO (search engine optimization) website data | |
CN104298785A (en) | Searching method for public searching resources | |
CN113297457B (en) | High-precision intelligent information resource pushing system and pushing method | |
CN103399855B (en) | Behavior intention determining method and device based on multiple data sources | |
CN105916032A (en) | Video recommendation method and video recommendation terminal equipment | |
CN102722499A (en) | Search engine and implementation method thereof | |
CN102737021A (en) | Search engine and realization method thereof |
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 |