CN108829793A - A kind of organizational member hobby method for digging - Google Patents
A kind of organizational member hobby method for digging Download PDFInfo
- Publication number
- CN108829793A CN108829793A CN201810558855.4A CN201810558855A CN108829793A CN 108829793 A CN108829793 A CN 108829793A CN 201810558855 A CN201810558855 A CN 201810558855A CN 108829793 A CN108829793 A CN 108829793A
- Authority
- CN
- China
- Prior art keywords
- interest
- item
- original
- organizational member
- hobby
- 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
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000004458 analytical method Methods 0.000 claims description 7
- 238000009412 basement excavation Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 230000008707 rearrangement Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005065 mining Methods 0.000 abstract description 3
- 238000012098 association analyses Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 244000097202 Rathbunia alamosensis Species 0.000 description 1
- 235000009776 Rathbunia alamosensis Nutrition 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001915 proofreading effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003997 social interaction Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Computing Systems (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of organizational member hobby method for digging, include the following steps：The higher multiple original item of interest of the frequency of occurrences are extracted in text from pushing away；It is ranked up according to the frequency of original item of interest, the frequency that original item of interest occurs is its original weight；The association item of interest set of high frequency item of interest is obtained according to interest association rule；Original item of interest is extracted one by one, if some original item of interest meets correlation rule and there are an association item of interest in association item of interest set, and the association item of interest is identical as another original item of interest, then increases the weight of another original item of interest；To treated, original item of interest re-starts sequence according to weight, item of interest of several original item of interest as the organizational member before extracting.The recall ratio and precision ratio of the hobby excavated by this method are improved, and mining effect is more preferable.
Description
Technical field
The present invention relates to data mining technology field more particularly to a kind of organizational member hobby method for digging.
Background technique
With the fast development of internet and universal, the use habit of internet organizational member oneself is found from initial
Content transforming is the content push for relying on service side and giving, therefore knows precisely each organizational member point of interest, can effectively be helped
It helps service side to provide personalized service, improves the usage experience of organizational member.
In terms of hobby excavation, Deng L etc. proposes a kind of to excavate China based on the algorithm of label and two-way interactive
The topic interest of the organizational member of one of maximum social interaction server Sina weibo.The algorithm passes through the system of organizational member interaction figure
It is fixed, it is sufficiently used the difference of the interaction between organizational member, the results showed that, the algorithm is in accuracy rate and recall rate side
Face is better than other methods, can effectively excavate organizational member to the interest of label and two-way interactive.
Vu T etc. constructs one and mentions from Twitter (website of U.S.'s social networks and microblog service) message
The system for taking organizational member interest, which extracts interested candidate item using language mode, and uses four kinds of different passes
Keyword ordering techniques are ranked up it：TF-IDF, Text Rank, LDA-Text Rank and RI-Rank.The result shows that TF-
IDF and Text Rank be suitable for from push away in text extract organizational member interest.
Bao H etc. proposes one based on time and social probability matrix decomposition model to predict organizational member in blog article
Potential interest.Model analysis temporal information and organizational member activity are to the potential feature space of organizational member and its interest master
The influence of topic provides the unified approach of time of fusion information and social network structure, following with Accurate Prediction organizational member
Interest.
The method of these internet organizational member interest diggings is based respectively on access log, the browsing content of microblogging or blog
And behavior.But the existing research work internal relation that is seldom related to hobby itself and these internal relations are in interest
Application in hobby excavation.
Summary of the invention
It is an object of the invention in view of the deficiencies in the prior art, propose organizational member hobby excavation side
Method.For the attainment of one's purpose, technical solution provided by the invention is：
A kind of organizational member hobby method for digging of the present invention, includes the following steps：
(1) multiple high frequency item of interest are sorted out from the Profile of all organizational members of social network sites, and from some group
Pushing away for the person of being made into excavates n item of interest identical with high frequency item of interest as original item of interest in text, n is the integer greater than 1；
(2) it according to the frequency of occurrence of original item of interest, sorts to original item of interest, forms original item of interest list, be denoted as
IttsSet 1~ittsSet n forms original item of interest set, the corresponding original item of interest of 1~ittsSet of ittsSet n
Initial weight w is respectively the frequency of occurrence that 1~w of w, 1~w of n, w n is corresponding original item of interest；
(3) according to the relevance of hobby Association Rule Analysis high frequency item of interest, association item of interest set is formed
1~ruleSet of ruleSet m, m are the integer greater than 1；
(4) original item of interest ittsSet x is extracted one by one, if it has association item of interest, which is present in
It is associated in item of interest set, is denoted as ruleSet y, and association item of interest ruleSet y and another original item of interest
IttsSet x ' is identical, then the weight for increasing original item of interest ittsSet x ' is W, x and x ' be any integer in 1~n and
It is not mutually equal, y is any integer in 1~m, and the weight of remaining original item of interest remains unchanged；
(5) according to the sequence of the new original item of interest of weight rearrangement, the organizational member item of interest list is obtained, is chosen
Item of interest of the maximum one or more item of interest of weight as the organizational member.
Preferably, emerging from the arrangement high frequency in all organizational member Profiles using segmentation methods in the step 1
Interesting item.
Preferably, the segmentation methods include the following steps：
(1.1) the interest word of description interest is sorted out from the Profile of all organizational members of social network sites；
(1.2) by artificial check and correction, it polymerize the interest word for expressing equivalent in meaning to form synonym collection, every group of synonym
A kind of corresponding item of interest of set；
(1.3) the interest word filled in social network sites by organizational member is replaced with into corresponding item of interest；
(1.4) frequency that all item of interest occur is calculated, and records multiple high frequency item of interest.
Preferably, in the step 3 excavation of hobby correlation rule method, include the following steps：
(3.1) minimal confidence threshold min_conf and minimum support threshold value min_sup is set；
(3.2) support is found out greater than minimum support threshold value min_sup, and confidence level is greater than minimal confidence threshold
Association item of interest of the associations of min_conf as original item of interest；
(3.3) the relevant item of interest of record institute forms association item of interest set.
Preferably, the minimal confidence threshold min_conf is 20%, and minimum support threshold value min_sup is
0.4%.
Preferably, the calculation that the weight of the step 4 Central Plains beginning item of interest ittsSet x ' is W be W=w+k ×
R, wherein parameter k is the constant for setting the capability of influence that correlation rule excavates hobby, and parameter r is original item of interest
IttsSet x is the probability of the true item of interest of the organizational member.
Using technical solution provided by the invention, compared with prior art, have the advantages that：
The present invention is by extracting vocabulary about hobby from pushing away for the network platform in text, these vocabulary of filing,
And analysis is associated to equivalent in meaning or similar vocabulary, increase the weight with relevance vocabulary, obtains more accurate group
The information of the person's of being made into hobby can promote the mining effect of the hobby method for digging based on word frequency, not be only social activity
Hobby excavation on network provides a new thinking, and interest digging effect is more preferable, discloses and is closing in hobby
Connection relationship and application value.
Detailed description of the invention
Attached drawing 1 is the flow chart of organizational member hobby method for digging of the present invention；
Attached drawing 2 is the recall ratio ratio that the hobby based on different value of K excavates；
Attached drawing 3 is the precision ratio ratio that the hobby based on different value of K excavates；
Attached drawing 4 is the F1 value ratio that the hobby based on different value of K excavates.
Specific embodiment
To further appreciate that the contents of the present invention, the present invention is described in detail in conjunction with the embodiments, and following embodiment is used for
Illustrate the present invention, but is not intended to limit the scope of the invention.
Embodiment one
In conjunction with shown in attached drawing 1, for the hobby of organizational member of the present embodiment to excavate the LinkedIn network platform,
Hobby method for digging of the present invention is illustrated, organizational member hobby method for digging packet of the present invention
Include following steps.
Step 1.1：LinkedIn organizational member would generally list their interest in Profile, the present embodiment from
Interest word is extracted in the Profile of all organizational members of the LinkedIn network platform to use first when data acquire
LinkedIn web crawlers, random collecting LinkedIn personal information sort out that retouch element emerging including acquiring their hobby
The interest word of interest.
Step 1.2：When LinkedIn organizational member creates Profile, there is no prepare one group of interest love by LinkedIn
It is good for selection, organizational member vocabulary used when filling in hobby be it is open, LinkedIn organizational member can be used
The interest of any lexical representation oneself, freely edits their hobby, and therefore, LinkedIn organizational member is filled in
Hobby is simultaneously lack of standardization.The interest word that statistics obtains is classified, is owned by way of manually proofreading by the step
The synonym collection of the synset of interest word, interest word corresponds to an item of interest.
Step 1.3：According to synonym collection, interest word will be filled in by organizational member oneself in organizational member Profile
Replace with item of interest.
Step 1.4：For each item of interest, the frequency and frequency of occurrence that each item of interest occurs in personal information are calculated
It is highest to extract the frequency of occurrences for indicating the generality of the item of interest with the percentage of total frequency of occurrence of all item of interest
As the high frequency item of interest of the present embodiment, (quantity of high frequency item of interest determines 210 item of interest according to the actual situation, this reality
Applying example is for extracting 210 high frequency item of interest).
Step 1.5：The present embodiment uses Twitter platform as the platform of analysis, and LinkedIn organizational member can be
It is created an account on Twitter platform, issues and read brief information, be known as pushing away text, and pushing away in text from some organizational member
N item of interest identical with high frequency item of interest are excavated as original item of interest, wherein n is the integer greater than 1.
Step 2：It according to the frequency of occurrence of original item of interest, sorts to original item of interest, forms original item of interest list, remember
For ittsSet 1~ittsSet n, original item of interest set, the corresponding original interest of 1~ittsSet of ittsSet n are formed
Item initial weight w is respectively the frequency of occurrence that 1~w of w, 1~w of n, w n is respectively corresponding original item of interest.
Step 3：According to the relevance of hobby Association Rule Analysis high frequency item of interest, association item of interest set is formed
RuleSet1~ruleSet m, m are the integer greater than 1, and specific steps include：
Step 3.1：Confidence level refers to the measurement of correlation rule accuracy and intensity, and the rule in data record collection D is set
Reliability refers to the frequency for occurring association item of interest ruleSet in all records for original item of interest ittsSet occur, also means
Rule certainty, be expressed as follows：
Support refers to the importance for measuring correlation rule, reflects the generality of correlation rule, and point out correlation rule
It is frequent degree in all record sets, regular grid DEM refers in all records while going out in data record collection D
Show original item of interest ittsSet and the frequency for being associated with item of interest ruleSet, is expressed as follows：
Minimal confidence threshold min_conf and minimum support threshold value min_sup, minimal confidence threshold are set first
Min_conf is 20%, and minimum support threshold value min_sup is 0.4%.
Step 3.2：Support is found out from numerous associations is greater than minimum support threshold value min_sup, and confidence
Association item of interest of associations of the degree greater than minimal confidence threshold min_conf as original item of interest.
Step 3.3：The relevant item of interest of record institute forms association item of interest set ruleSet1~ruleSet m, and m is big
In 1 integer.
Step 4：Original item of interest ittsSet x is extracted one by one, if it meets correlation rule and in association item of interest set
It is middle that there are association an item of interest ruleSet y, association item of interest ruleSet y and another original item of interest ittsSet
X ' is identical, then the weight for increasing original item of interest ittsSet x ' is W, x and x ' it is any integer in 1~n and is not mutually equal,
Y is any integer in 1~m, and the calculation of original item of interest ittsSet x ' new weight W is W=w+k × r, wherein
Parameter k is the constant for setting the capability of influence that correlation rule excavates hobby, and parameter r is original item of interest ittsSet
X is the probability of the true item of interest of the organizational member, and the weight of remaining original item of interest remains unchanged.
Step 5：According to the sequence of the new original item of interest of weight rearrangement, the organizational member item of interest list is obtained,
Item of interest of the maximum one or more item of interest of weight selection as the organizational member.
Effect example：
After being reordered based on organizational member item of interest list, obtained the results list, the present embodiment using precision ratio,
Recall ratio and F1 value analyze mining effect as the evaluation index of model, and recall ratio refers to the item of interest excavated and true
The percentage of real item of interest, precision ratio refer to true item of interest and the percentage of all item of interest excavated, and F1 value refers to and looks into
The average value of full rate and precision ratio, their definition difference are as follows：
Wherein, TP indicates the quantity for the true item of interest of the item of interest person of being organized into excavated, and FN expression is excavated emerging
The quantity of the interesting item not true item of interest of the person of being organized into, FP indicate the quantity for the true item of interest of organizational member that do not excavate.
The weight calculation mode of the original item of interest of some obtained after step 4 is W=w+k × r, it is known that as parameter k
When being arranged to 0, correlation rule does not work, and method only returns to the original orderly interest list based on word frequency, wherein interest
The sequence of item is the frequency of occurrence pushed away based on them in organizational member in text, i.e., without obtaining organizational member by association analysis
Item of interest.
The orderly item of interest list of computation organization member after reordering n-th item of interest person of being organized into it is true emerging
The probability of interest hobby, i.e., the hit rate of n-th item of interest.For example, in the orderly item of interest list of each organizational member
First item of interest, when parameter k is respectively set to 0,7,14,21,28,35 and 42, corresponding hit rate is respectively
17.42%, 19.25%, 21.61%, 21.72%23.23%, 23.44% and 23.76%；And for the second of organizational member
A item of interest, corresponding ratio are respectively 15.32%, 16.40%, 16.18%, 17.69%, 17.04%, 17.26% He
16.94%, that is, work as k=7, when 14,21,28,35 or 42, hit rate when hit rate is compared with k=0 is all high.It can be seen that through reaching a standard
The hit rate of the organizational member hobby obtained after connection analysis is to be promoted.
Choose some organizational member, preceding 10 item of interest in his orderly interest list available first simultaneously calculate him
Recall ratio, organizational member quantity accounting when then calculating its recall ratio greater than a certain given value.When parameter k is respectively 0,7
When with 42, organizational member ratio of the recall ratio greater than 70% is respectively 6.77%, 7.96% and 8.17%.Recall ratio is greater than 30%
Organizational member ratio be respectively 40.22%, 45.59% and 48.49%.As shown in Fig. 2, it can intuitively illustrate and work as parameter
Recall ratio ratio chart when k is respectively 0,7 and 42.It can be seen that the organizational member item of interest obtained after association analysis
Recall ratio gets a promotion.
Similarly, some organizational member is chosen, preceding 10 interest in he same available orderly item of interest list
Then the precision ratio and F1 value of item calculate the organizational member quantity accounting of its precision ratio and F1 value greater than a certain given value when.Such as
Attached drawing 3 and attached drawing 4 respectively show the ratio chart of precision ratio and F1 value when parameter k value takes 0,7 and 42.By attached drawing 3 and attached
When the value of parameter k is 0 known to Fig. 4, corresponding curve is not so good as other curves, and the curve that the value for corresponding to parameter k is 42 is then
It is best.This also means that association analysis, which has hobby excavation, promotes effect.
It describes the invention in detail in conjunction with the embodiments above, but the content is only preferable implementation of the invention
Example, should not be considered as limiting the scope of the invention.It is all according to all the changes and improvements made by the present patent application range
Deng should all still fall within patent covering scope of the invention.
Claims (6)
1. a kind of organizational member hobby method for digging, which is characterized in that include the following steps：
(1) multiple high frequency item of interest, and pushing away in text from some organizational member are sorted out from all organizational member Profiles
N item of interest identical with high frequency item of interest are excavated as original item of interest, n is the integer greater than 1；
(2) it according to the frequency of occurrence of original item of interest, sorts to original item of interest, forms original item of interest list, be denoted as
IttsSet 1~ittsSet n forms original item of interest set, the corresponding original item of interest of 1~ittsSet of ittsSet n
Initial weight w is respectively w1~wn, and w1~wn is the frequency of occurrence of corresponding original item of interest；
(3) according to the relevance of hobby Association Rule Analysis high frequency item of interest, association item of interest set ruleSet 1 is formed
~ruleSet m, m are the integer greater than 1；
(4) original item of interest ittsSet x is extracted one by one, if it has association item of interest, which is present in association
In item of interest set, it is denoted as ruleSet y, and association item of interest ruleSet y and another original item of interest ittsSet
X ' is identical, then the weight for increasing original item of interest ittsSet x ' is W, x and x ' it is any integer in 1~n and is not mutually equal,
Y is any integer in 1~m, and the weight of remaining original item of interest remains unchanged；
(5) according to the sequence of the new original item of interest of weight rearrangement, the organizational member item of interest list, weight selection are obtained
Maximum one or more item of interest is the item of interest of the organizational member.
2. organizational member hobby method for digging according to claim 1, it is characterised in that：It is adopted in the step 1
With segmentation methods from the arrangement high frequency item of interest in all organizational member Profiles.
3. organizational member hobby method for digging according to claim 2, which is characterized in that the segmentation methods packet
Include following steps：
(1.1) the interest word of description interest is sorted out from the Profile of all organizational members of social network sites；
(1.2) by artificial check and correction, it polymerize the interest word for expressing equivalent in meaning to form synonym collection, every group of synonym collection
A kind of corresponding item of interest；
(1.3) the interest word filled in social network sites by organizational member is replaced with into corresponding item of interest；
(1.4) frequency that all item of interest occur is calculated, and records multiple high frequency item of interest.
4. organizational member hobby method for digging according to claim 1, which is characterized in that emerging in the step 3
The method of the excavation of interest hobby correlation rule, includes the following steps：
(3.1) minimal confidence threshold min_conf and minimum support threshold value min_sup is set；
(3.2) support is found out greater than minimum support threshold value min_sup, and confidence level is greater than minimal confidence threshold min_
Association item of interest of the associations of conf as original item of interest；
(3.4) the relevant item of interest of record institute forms association item of interest set.
5. organizational member hobby method for digging according to claim 4, it is characterised in that：The min confidence
Threshold value min_conf is 20%, and minimum support threshold value min_sup is 0.4%.
6. organizational member hobby method for digging according to claim 1, which is characterized in that step 4 Central Plains
The calculation that the weight of beginning item of interest ittsSet x ' is W is W=w+k × r, wherein parameter k be setting correlation rule for
The constant for the capability of influence that hobby excavates, parameter r are that original item of interest ittsSet x is the true interest of the organizational member
The probability of item.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810558855.4A CN108829793B (en) | 2018-06-01 | 2018-06-01 | Organization hobby mining method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810558855.4A CN108829793B (en) | 2018-06-01 | 2018-06-01 | Organization hobby mining method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108829793A true CN108829793A (en) | 2018-11-16 |
| CN108829793B CN108829793B (en) | 2021-09-24 |
Family
ID=64147148
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810558855.4A Active CN108829793B (en) | 2018-06-01 | 2018-06-01 | Organization hobby mining method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108829793B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110647676A (en) * | 2019-08-14 | 2020-01-03 | 平安科技（深圳）有限公司 | Interest attribute mining method and device based on big data and computer equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020143759A1 (en) * | 2001-03-27 | 2002-10-03 | Yu Allen Kai-Lang | Computer searches with results prioritized using histories restricted by query context and user community |
| US20130031088A1 (en) * | 2010-02-10 | 2013-01-31 | Python4Fun, Inc. | Finding relevant documents |
| CN104217013A (en) * | 2014-09-22 | 2014-12-17 | 广西教育学院 | Course positive and negative mode excavation method and system based on item weighing and item set association degree |
| CN105740381A (en) * | 2016-01-27 | 2016-07-06 | 北京工业大学 | User interest mining method based on complex network characteristics and neural network clustering |
| CN105912673A (en) * | 2016-04-11 | 2016-08-31 | 天津大学 | Optimization method for Micro Blog search based on personalized characteristics of user |
| CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
-
2018
- 2018-06-01 CN CN201810558855.4A patent/CN108829793B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020143759A1 (en) * | 2001-03-27 | 2002-10-03 | Yu Allen Kai-Lang | Computer searches with results prioritized using histories restricted by query context and user community |
| US20130031088A1 (en) * | 2010-02-10 | 2013-01-31 | Python4Fun, Inc. | Finding relevant documents |
| CN104217013A (en) * | 2014-09-22 | 2014-12-17 | 广西教育学院 | Course positive and negative mode excavation method and system based on item weighing and item set association degree |
| CN105740381A (en) * | 2016-01-27 | 2016-07-06 | 北京工业大学 | User interest mining method based on complex network characteristics and neural network clustering |
| CN105912673A (en) * | 2016-04-11 | 2016-08-31 | 天津大学 | Optimization method for Micro Blog search based on personalized characteristics of user |
| CN107633430A (en) * | 2017-09-20 | 2018-01-26 | 哈尔滨工业大学 | A kind of Method of Commodity Recommendation based on community of colony |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110647676A (en) * | 2019-08-14 | 2020-01-03 | 平安科技（深圳）有限公司 | Interest attribute mining method and device based on big data and computer equipment |
| CN110647676B (en) * | 2019-08-14 | 2023-04-11 | 平安科技（深圳）有限公司 | Interest attribute mining method and device based on big data and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108829793B (en) | 2021-09-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210209109A1 (en) | Method, apparatus, device, and storage medium for intention recommendation | |
| US8332775B2 (en) | Adaptive user feedback window | |
| US10776431B2 (en) | System and method for recommending content based on search history and trending topics | |
| US11122009B2 (en) | Systems and methods for identifying geographic locations of social media content collected over social networks | |
| CN103049440B (en) | A kind of recommendation process method of related article and disposal system | |
| US20140143405A1 (en) | System And Method For Analyzing Social Media Trends | |
| US20150032492A1 (en) | Methods of Identifying Relevant Content and Subject Matter Expertise for Online Communities | |
| KR102080362B1 (en) | Query expansion | |
| CN103593413B (en) | META Search Engine personalized method based on Agent | |
| Hahmann et al. | Twitter location (sometimes) matters: Exploring the relationship between georeferenced tweet content and nearby feature classes | |
| CN103823888B (en) | Node-closeness-based social network site friend recommendation method | |
| RU2670494C2 (en) | Method for processing search requests, server and machine-readable media for its implementation | |
| US20130297694A1 (en) | Systems and methods for interactive presentation and analysis of social media content collection over social networks | |
| US20130297581A1 (en) | Systems and methods for customized filtering and analysis of social media content collected over social networks | |
| CN105045931A (en) | Video recommendation method and system based on Web mining | |
| EP2407897A1 (en) | Device for determining internet activity | |
| CN104008203A (en) | User interest discovering method with ontology situation blended in | |
| CN104657425A (en) | Topic management type network public opinion evaluation management system and method | |
| CN104866557B (en) | A kind of personalized instant learning theoretical based on constructive learning supports System and method for | |
| CN102609475A (en) | Method for monitoring content of microblog and monitoring system | |
| CN103020302A (en) | Academic core author excavation and related information extraction method and system based on complex network | |
| CN103268350A (en) | Internet public opinion information monitoring system and monitoring method | |
| CN104077417A (en) | Figure tag recommendation method and system in social network | |
| CN103473036B (en) | A kind of input method skin method for pushing and system | |
| CN104899236B (en) | A kind of comment information display methods, apparatus and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Zhou Jiayong Inventor after: Si Huayou Inventor after: Wan Jian Inventor after: Chen Zhihui Inventor after: Wu Haopeng Inventor after: Sun Wen Inventor before: Zhou Jiayong Inventor before: Si Huayou Inventor before: Chen Zhihui Inventor before: Wu Haopeng Inventor before: Sun Wen |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |