CN103886074B - Commercial product recommending system based on social media - Google Patents
Commercial product recommending system based on social media Download PDFInfo
- Publication number
- CN103886074B CN103886074B CN201410110393.1A CN201410110393A CN103886074B CN 103886074 B CN103886074 B CN 103886074B CN 201410110393 A CN201410110393 A CN 201410110393A CN 103886074 B CN103886074 B CN 103886074B
- Authority
- CN
- China
- Prior art keywords
- user
- commodity
- social media
- commercial product
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000013065 commercial product Substances 0.000 title claims abstract description 52
- 238000013507 mapping Methods 0.000 claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 20
- 238000004458 analytical method Methods 0.000 claims abstract description 17
- 230000004927 fusion Effects 0.000 claims abstract description 14
- 238000013481 data capture Methods 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 16
- 238000010801 machine learning Methods 0.000 claims description 13
- 230000002452 interceptive effect Effects 0.000 claims description 12
- 230000003993 interaction Effects 0.000 claims description 6
- 238000005295 random walk Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000010219 correlation analysis Methods 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000003672 processing method Methods 0.000 claims description 3
- 230000000644 propagated effect Effects 0.000 claims description 3
- 238000006748 scratching Methods 0.000 claims description 3
- 230000002393 scratching effect Effects 0.000 claims description 3
- 230000010354 integration Effects 0.000 abstract 1
- 239000000047 product Substances 0.000 description 6
- 241000270322 Lepidosauria Species 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000001902 propagating effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000003542 behavioural effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000008451 emotion Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010297 mechanical methods and process Methods 0.000 description 1
- 230000005226 mechanical processes and functions Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
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
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0269—Targeted advertisements based on user profile or attribute
- G06Q30/0271—Personalized advertisement
-
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Finance (AREA)
- Economics (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of commercial product recommending system based on social media, including social media data capture module, electronic commerce data handling module, information extraction, fusion, comparison module and commercial product recommending module；Social media data capture module carries out data grabber for social media；Electronic commerce data handling module carries out data grabber for e-commerce website；Information extraction, fusion, comparison module are used for the data that crawl is obtained to be analyzed, processed, and commodity are carried out with the foundation and mapping of commodity tree, analysis is modeled to social media user.Commercial product recommending module carries out merchandise query for treating recommended users, is recommended according to degree of association, interest, circle of friends information.The present invention combines an integration system of marketing for social media and ecommerce, by social media and ecommerce, realizes that commercial product recommending, the commodity purchasing interest of user fully can be excavated under the conditions of without client's subjectivity row.
Description
Technical field
The present invention relates to cyber application, specifically a kind of commercial product recommending system based on social media.
Background technology
With the development and popularization of Web2.0, social networkies are increasingly received by numerous Internet users.User utilizes
The Internet carries out doings, promotes oneself, shows emotion, or makes friends.All of activity on social networkies is all
Illustrate the interest and preference of user.In recent years, the flow of social networkies is cashed and is increasingly becoming Internet advertising and recommendation
The Forward researches of system.Success to user's Recommendations and is efficiently converted into click and then buys, and can form similar search
The advertisement Internet industry scale similar with the display advertisement of door network.Therefore, to the user's Recommendations in social media
It is very actual problem.
However, Recommendations are the problems of a relative difficult on social networkies.For example, user delivers word and figure
Piece content is not perhaps directly contacted with commodity（User likes the picture for sending out famous mountains and great rivers, and the picture of tourism）If given
He recommends some commodity（Such as outdoor products）Which will not be at least made to dislike, it could even be possible to triggering is clicked on.Therefore, how use is found
The interest at family, and can excavate and commodity in use between relation be this patent problem to be solved..
Prior art is generally basede on the Collaborative Recommendation of friend or the similarity based on content understanding is recommended.
Collaborative Recommendation based on friend generally needs the hobby for knowing friend, such as A to have 20 friends, wherein have 10 people
Like film, then can predict that A is likely to like film.Advantage of this is that the letter for taking full advantage of social networkies
Breath, and be more accurately based on the recommendation of the hobby of friend.However, on a lot of social networkies, not friend's purchase
Or the information of the commodity of concern.Therefore, the problem for being easy to run into " cold start-up " based on the recommendation of friend, that is, have no idea to start
Prediction.Highly useful signal can be become based on the recommendation of friend, the commercial product recommending and concern on and if only if social networkies
Formed just more effective during certain scale.
Referred to based on the recommendation of content understanding and the article content that social network user is delivered is understood, and then estimate which
Preference.In addition, the label information on social networkies can be used to the preference for describing user.These preferences can be used to mate business
Product.For example, certain user has label " open air ", then outdoor products just become alternative Recommendations well.This method
Although can be used to the preference for estimating user, user can not be described in fine granularity well.For example, it is recommended that family
During outer product, it is row mountain on earth, or dives under water；Or be expensive anorak on earth, still relatively relatively inexpensive slide plate.These
Problem is judged using label is all more difficult.
For example, Patent publication No CN102479366 discloses a kind of Method of Commodity Recommendation and system, by obtaining user's
Behavioral data determines that the interest commodity of user tire out, and lifts shopping impression, but, this application has to rely on the behavior number with user
According to, i.e. click, search behavior data of the user in website, the commercial product recommending after client's subjectivity behavior lag behind customer action,
The commodity purchasing interest of user is not fully excavated.
For example, Patent publication No CN102592223 discloses a kind of Method of Commodity Recommendation and commercial product recommending system, according to use
Browsing for family is recorded or user property acquisition sample training data, and commercial product recommending still lags behind customer action.
At present, also there is no a kind of commending system fully based on social media information, the commodity purchase of user in prior art
Thing is not fully excavated.
Content of the invention
For solving prior art problem, the invention provides a kind of using various information in social media（Content, network
Deng）, and commodity are effectively understood, form the commercial product recommending system of a system.
The technical scheme is that：A kind of commercial product recommending system based on social media, grabs including social media data
Delivery block, electronic commerce data handling module, information extraction, fusion, comparison module and commercial product recommending module；
Social media data capture module carries out data grabber for social media, carries out multimachine by parallel algorithms
Crawl data；
Electronic commerce data handling module carries out data grabber for e-commerce website, is carried out by parallel algorithms
Multimachine captures data, carries out comprehensive data interaction by DeepWeb query expansions；
Information extraction, fusion, comparison module are used for the data that crawl is obtained to be analyzed, processed, and commodity are entered to do business
The foundation and mapping of product tree, is modeled analysis to social media user.
Commercial product recommending module carries out merchandise query for treating recommended users, the commodity for obtaining is ranked up, according to phase
Guan Du, interest, circle of friends information are recommended；
Social media data capture module, electronic commerce data handling module, commercial product recommending module with information extraction, melt
Close, comparison module is connected.
Social media data capture module obtains the data in social media by web crawlers, after obtaining data, passes through
The task that a plurality of URL is captured is distributed to multiple stage computers using hadoop so that the scheduling of the load balancing of every computer
Processing method gives the distributed system constituted by multi-section server, webpage is analyzed by HTML parser, text point
The control of analysis, link analysis and web page quality, duplicate removal, obtain corresponding web page contents, and web page contents result is divided into structuring letter
Breath（The link informations such as friend, group）And unstructured information（Text, image etc.）, be respectively stored into structured message data base and
In the Unstructured Information Database.By distributed system, the information of very high-throughput can be processed.
Structuring and non-structured classification can be by judging whether the content is possibly stored in structured database
（Such as SQL）To judge.Generally text and image are unstructured datas, it is impossible to which content therein is carried out cutting and classification.Such as
One section of news, although know that there are the information such as name, place name, exabyte, time the inside, but if do not processed, it is impossible to from
Dynamic imports to these information in SQL.
Electronic commerce data handling module is a series of for e-commerce website is produced by machine learning Query Builder
Regular expression query statement is inquired about to e-commerce website, and all of information scratching is got off, and is looked into by DeepWeb
Extension, page analyzer are ask carrying out the data interaction of web page contents.The inquiry that different web sites are learnt by machine learning algorithm
Rule, is traveled through all inquiries with maximum probability using the method that key word is replaced, is intelligently inquired about by knowledge base, item property
Extract, true value finds, generate entity attribute relational database, it is also possible to train grader to carry out by historical data.
Information extraction, fusion, comparison module include that user modeling module, commodity MBM, commodity and user mapping is built
Mould module, user modeling module, commodity MBM are connected with commodity and user's mapping, modeling module.
User modeling module routine includes：
1-1）Label is propagated：For social network user, by the label on social networkies（Label on social networkies
It can be the label of user oneself mark）The figure of composition carries out the probability that random walk obtains label propagation, so as to extending user
Label；
1-2）Content differentiates：The content that user delivers is analyzed, is obtained using topic model, entity extraction possible
Label；Meanwhile, the user of existing label is learnt by training machine Study strategies and methods, so as to the user without label
Enter row label judgement；
1-3）User's other information differentiates：For the content that user delivers is understood and its circle of friends is carried out point
Analysis, and then age of user, job specification, job site, income information is estimated, such that it is able to be better understood from the need of user
Ask；When estimating to the age of user, job specification, job site and income information, key word and good friend are extracted to user
Attribute character, using machine learning method, carries out study to existing markup information and obtains grader, unknown sample is carried out point
Class.
It is first to be estimated to each user an age to the analysis of circle of friends, this age can be that he fills in social activity
Age on network, or we by initializing a model content that he delivers being returned obtained by year
Age.And then the process for carrying out propagating similar to label on social networkies, the circle of friends of the user is analyzed, friend is obtained
Age bracket statistics, from row change the user estimation of Age.
Commodity MBM running includes,
2-1）Property value is filled, and the item property obtained by webpage capture may be needed based on the Internet not enough comprehensively
Search engine is scanned for, and obtains possible property value from corresponding summary and ad content, and counts the probability of appearance,
Make a look up in the Internet, mate, counting the frequency of occurrences to carry out true value discovery；
2-2）In addition commodity tree classification, obtain taxonomy-tree information to without classification tree information simultaneously in crawl commodity
Commodity are classified, and are categorized on some node of commodity tree；
2-3）Commodity other information is gathered, and the other information of commodity is collected, and is stored in data base, by right
The structure of e-commerce website is analyzed, and is commented on accordingly and is given a mark.
Commodity are corresponding with user's foundation mapping by commodity with user's mapping, modeling module, and user is mentioned in social media
Comment denoising of the picture, user in the content=that delivers during dependent merchandise and e-commerce website to commodity and businessman, sets up
Mapping model, mapping model are the direct feature extraction to data or the mark sheet obtained by the means of machine learning
Reach, after having obtained mapping model, compare the dependency of commodity and user.
Commercial product recommending module includes that recommended by client module and commodity end recommending module, recommended by client module are used for use
Family Recommendations, commodity end recommending module are used for commercial product recommending user, and recommended by client module, commodity end recommending module are equal
With information extraction, fusion, comparison module are connected.
Recommended by client module running is comprised the following steps：
3-1）The items list that may recommend is obtained by the dependency of user and commodity；
3-2）The analysis of the dependency of user and commodity is carried out to the good friend of user, and by the items list of good friend to this
User is voted；
3-3）Commercial product recommending is further processed by analyzing user's portrait, target, user's portrait bag are recommended in subdivision
Include age, income and interest；User's portrait refers to the related content of user modeling module；
3-4）Commercial product recommending is carried out for the user by the interactive mode of social media, the interactive mode of social media includes
Add good friend, quote good friend, personal letter, comment etc..
Commercial product recommending module running is comprised the following steps：
4-1）By the dependency of commodity and user obtain may be interested in the commodity user；
4-2）The correlation analysiss of commodity and user are carried out to the good friend of user, and by the items list of good friend to the use
Voted at family；
4-3）Commercial product recommending is further processed by analyzing user's portrait, target, user's portrait bag are recommended in subdivision
Include age, income and interest；
4-4）Commercial product recommending is carried out for the user by the interactive mode of social media, the interactive mode of social media includes
Add good friend, quote good friend, personal letter, comment etc..
Beneficial effect of the present invention includes that the present invention integrates system for one that social media and ecommerce combine marketing
System, by social media and ecommerce, realizes commercial product recommending under the conditions of without client's subjectivity row, and the commodity purchasing of user is emerging
Interest fully can be excavated.
Further, the system effectively make use of the characteristics of social media and ecommerce, due to each at present big social activity
Media platform and e-commerce platform are all actively releasing payment UNICOM business, and it will be wide therefore to provide third-party marketing system
Big advertiser（Electric business）There is provided effectively automatically or semi-automatically mode come carry out goods marketing, businessman promote and advanced level user cash
Provide more approach.
Description of the drawings
Fig. 1 is the structural representation of the present invention；
Fig. 2 is that social media data capture module process processes schematic diagram；
Fig. 3 is electronic commerce data handling module processing procedure schematic diagram；
Fig. 4 is information extraction, fusion, comparison module structural representation.
Specific embodiment
Technical solution of the present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings, so that ability
The technical staff in domain can be better understood from the present invention and can be practiced, but illustrated embodiment is not as the limit to the present invention
Fixed.
As shown in figure 1, a kind of commercial product recommending system based on social media, including social media data capture module, electricity
Sub- business data handling module, information extraction, fusion, comparison module and commercial product recommending module；
Social media data capture module carries out data grabber for social media, carries out multimachine by parallel algorithms
Crawl data；
Electronic commerce data handling module carries out data grabber for e-commerce website, is carried out by parallel algorithms
Multimachine captures data, carries out comprehensive data interaction by DeepWeb query expansions；
Information extraction, fusion, comparison module are used for the data that crawl is obtained to be analyzed, processed, and commodity are entered to do business
The foundation and mapping of product tree, is modeled analysis to social media user.
Commercial product recommending module carries out merchandise query for treating recommended users, the commodity for obtaining is ranked up, according to phase
Guan Du, interest, circle of friends information are recommended；
Social media data capture module, electronic commerce data handling module, commercial product recommending module with information extraction, melt
Close, comparison module is connected.
As shown in Fig. 2 social media data capture module passes through web crawlers（Including N number of reptile, reptile 1, reptile 2, climb
Worm 3 ... reptile N）Obtain social media on data, after obtaining data, by using hadoop by a plurality of URL capture times
Business allocation schedule is processed to multiple stage computers so that the scheduling processing method of the load balancing of every computer gives multi-section service
The distributed system constituted by device, webpage is analyzed by HTML parser, text analyzing, link analysis and webpage matter
Amount control, duplicate removal, obtain corresponding web page contents, the web page contents result are divided into structured message（Friend, group etc. link
Information）And unstructured information（Text, image etc.）, it is respectively stored into structured message data base and unstructured information data
In storehouse.
Structuring and non-structured classification can be by judging whether the content is possibly stored in structured database
（Such as SQL）To judge.Generally text and image are unstructured datas, it is impossible to which content therein is carried out cutting and classification.Such as
One section of news, although know that there are the information such as name, place name, exabyte, time the inside, but if do not processed, it is impossible to from
Dynamic imports to these information in SQL.Meanwhile, structured message and unstructured information can also repeat webpage and carry out
The control of analysis, text analyzing, link analysis and web page quality, duplicate removal, obtain structured message and the unstructured information that simplifies.
As shown in figure 3, electronic commerce data handling module passes through machine learning Query Builder for e-commerce website
Produce a series of regular expression query statements to inquire about e-commerce website, and all of information scratching is got off, lead to
Cross DeepWeb query expansions, page analyzer to carry out the data interaction of web page contents.Different by machine learning algorithm study
The rule searching of website, is traveled through all inquiries with maximum probability using the method that key word is replaced, is intelligently looked into by knowledge base
Ask, item property is extracted, true value finds, generate entity attribute relational database, it is also possible to by historical data training point
Class device is carried out.
Regular expression query statement for example, as it is known that the inquiry of a Taobao be " man, light, 40, running shoe, nike are interior
Color ", can be extended by knowledge base, for example size, color, type, brand, then can grab a greater variety of footwear,
Crawl is being realized by analyzing related pages.
As shown in figure 4, information extraction, fusion, comparison module include user modeling module, commodity MBM, commodity and
User's mapping, modeling module, user modeling module, commodity MBM are connected with commodity and user's mapping, modeling module.
User modeling module routine includes：
1-1）Label is propagated：For social network user, by the label on social networkies（Label on social networkies
It can be the label of user oneself mark）The figure of composition carries out the probability that random walk obtains label propagation, the side of random walk
Formula is the good friend's network random walk consisted of user, so as to the label of extending user；
1-2）Content differentiates：The content that user delivers is analyzed, is obtained using topic model, entity extraction possible
Label；Meanwhile, the user of existing label is learnt by training machine Study strategies and methods, so as to the user without label
Enter row label judgement；The method that topic model refers to a class machine learning, the present embodiment use Latent Dirichlet
Allocation, in actual mechanical process, can be not limited to using this kind of method, it might even be possible to using text cluster or
Directly a topic is represented using high-frequency key words.
1-3）User's other information differentiates：For the content that user delivers is understood and its circle of friends is carried out point
Analysis, and then age of user, job specification, job site, income information is estimated, such that it is able to be better understood from the need of user
Ask；When estimating to the age of user, job specification, job site and income information, key word and good friend are extracted to user
Attribute character, using machine learning method, carries out study to existing markup information and obtains grader, unknown sample is carried out point
Class.
It is first to be estimated to each user an age to the analysis of circle of friends, this age can be that he fills in social activity
Age on network, or we by initializing a model content that he delivers being returned obtained by year
Age.And then the process for carrying out propagating similar to label on social networkies, the circle of friends of the user is analyzed, friend is obtained
Age bracket statistics, from row change the user estimation of Age.
Commodity MBM running includes,
2-1）Property value is filled, and the item property obtained by webpage capture may be needed based on the Internet not enough comprehensively
Search engine is scanned for, and obtains possible property value from corresponding summary and ad content, and counts the probability of appearance,
Make a look up in the Internet, mate, counting the frequency of occurrences to carry out true value discovery.
2-2）In addition commodity tree classification, obtain taxonomy-tree information to without classification tree information simultaneously in crawl commodity
Commodity are classified, and are categorized on some node of commodity tree；
2-3）Commodity other information is collected, and the other information of commodity is collected, and is stored in data base, by right
The structure of e-commerce website is analyzed, and is commented on accordingly and is given a mark.
Commodity are corresponding with user's foundation mapping by commodity with user's mapping, modeling module, and user is mentioned in social media
Comment denoising of the picture, user in the content=that delivers during dependent merchandise and e-commerce website to commodity and businessman, sets up
Mapping model, mapping model are the direct feature extraction to data or the mark sheet obtained by the means of machine learning
Reach, after having obtained mapping model, compare the dependency of commodity and user.
Commercial product recommending module includes recommended by client module and commodity end recommending module.
Recommended by client module running is comprised the following steps：
3-1）The items list that may recommend is obtained by the dependency of user and commodity；
3-2）The analysis of the dependency of user and commodity is carried out to the good friend of user, and by the items list of good friend to this
User is voted；
3-3）Commercial product recommending is further processed by analyzing user's portrait, subdivision recommends target, the user to draw
As including age, income and interest；
3-4）Commercial product recommending, the interactive mode of the social media are carried out for the user by the interactive mode of social media
Including adding good friend, quoting good friend, personal letter, comment etc.；
Commercial product recommending module running is comprised the following steps：
4-1）By the dependency of commodity and user obtain may be interested in the commodity user；
4-2）The correlation analysiss of commodity and user are carried out to the good friend of user, and by the items list of good friend to the use
Voted at family；
4-3）Commercial product recommending is further processed by analyzing user's portrait, subdivision recommends target, the user to draw
As including age, income and interest；
4-4）Commercial product recommending, the interactive mode of the social media are carried out for the user by the interactive mode of social media
Including adding good friend, quoting good friend, personal letter, comment etc..
The above is only the preferred embodiment of the present invention, it should be pointed out that：For those skilled in the art come
Say, under the premise without departing from the principles of the invention, can also make some improvements and modifications, these improvements and modifications also should be regarded as
Protection scope of the present invention.
Claims (9)
1. a kind of commercial product recommending system based on social media, it is characterised in that including social media data capture module, electronics
Business data handling module, information extraction, fusion, comparison module and commercial product recommending module；
Social media data capture module carries out data grabber for social media, carries out multimachine crawl by parallel algorithms
Data；Electronic commerce data handling module carries out data grabber for e-commerce website, is carried out by parallel algorithms many
Machine captures data, carries out comprehensive data interaction by DeepWeb query expansions；
Information extraction, fusion, comparison module are used for the data that crawl is obtained to be analyzed, processed, and carry out commodity tree to commodity
Foundation and mapping, analysis is modeled to social media user；
Commercial product recommending module is used for treating recommended users and carrying out merchandise query, and the commodity for obtaining are ranked up, according to degree of association,
Interest, circle of friends information are recommended；
Social media data capture module, electronic commerce data handling module, commercial product recommending module with information extraction, fusion,
Comparison module is connected；
The electronic commerce data handling module is a series of for e-commerce website is produced by machine learning Query Builder
Regular expression query statement is inquired about to e-commerce website, and all of information scratching is got off, and is looked into by DeepWeb
Extension, page analyzer are ask carrying out the data interaction of web page contents；The inquiry that different web sites are learnt by machine learning algorithm
Rule, is traveled through all inquiries with maximum probability using the method that key word is replaced, is intelligently inquired about by knowledge base, item property
Extract, true value finds, generate entity attribute relational database.
2. the commercial product recommending system based on social media according to claim 1, it is characterised in that the social media number
The data in social media are obtained according to handling module by web crawlers, after obtaining data, by using hadoop by a plurality of URL
The task of crawl distributes to multiple stage computers so that the scheduling processing method of the load balancing of every computer gives multi-section service
The distributed system constituted by device, webpage is analyzed by HTML parser, text analyzing, link analysis and webpage matter
Amount control, duplicate removal, obtain corresponding web page contents, and the web page contents result is divided into structured message and destructuring letter
Breath, is respectively stored in structured message data base and the Unstructured Information Database.
3. the commercial product recommending system based on social media according to claim 1, it is characterised in that described information extracts,
Fusion, comparison module include user modeling module, commodity MBM, commodity and user's mapping, modeling module, user modeling mould
Block, commodity MBM are connected with commodity and user's mapping, modeling module.
4. the commercial product recommending system based on social media according to claim 3, it is characterised in that the user modeling mould
The block course of work includes：
1-1) label is propagated：For social network user, random walk is carried out by the figure of the label composition on social networkies
The probability of label propagation is obtained, so as to the label of extending user；
1-2) content differentiates：The content that user delivers is analyzed, is extracted using topic model, entity and is obtained possible mark
Sign；Meanwhile, the user of existing label is learnt by training machine Study strategies and methods, so as to enter to the user without label
Row label judges；
1-3) user's other information differentiates：For the content that user delivers is understood and its circle of friends is analyzed, enter
And age of user, job specification, job site, income information is estimated, such that it is able to be better understood from the demand of user；To with
When the age at family, job specification, job site and income information are estimated, key word is extracted to user and good friend's attribute is special
Levy, using machine learning method, study is carried out to existing markup information and obtains grader, unknown sample is classified.
5. the commercial product recommending system based on social media according to claim 3, it is characterised in that the commodity model mould
Block running includes,
2-1) property value filling, the item property obtained by webpage capture may be needed based on internet hunt not enough comprehensively
Engine is scanned for, and obtains possible property value from corresponding summary and ad content, and counts the probability of appearance, in interconnection
Make a look up in net, mate, counting the frequency of occurrences to carry out true value discovery；
In addition 2-2) commodity tree classification, obtain taxonomy-tree information to the commodity without classification tree information simultaneously in crawl commodity
Classified, and be categorized on some node of commodity tree；
2-3) collection of commodity other information, is collected to the other information of commodity, and stores in data base, by electronics
The structure of business web site is analyzed, and is commented on accordingly and is given a mark.
6. the commercial product recommending system based on social media according to claim 3, it is characterised in that the commodity and user
It is corresponding that commodity and user are set up mapping by mapping, modeling module, deliver when mentioning dependent merchandise to user in the social media in
Comment denoising of the picture, user in appearance and e-commerce website to commodity and businessman, sets up mapping model, and mapping model is
The feature representation obtained to the direct feature extraction of data or by the means of machine learning, after having obtained mapping model,
Compare the dependency of commodity and user.
7. the commercial product recommending system based on social media according to claim 1, it is characterised in that the commercial product recommending mould
Block includes recommended by client module and commodity end recommending module.
8. the commercial product recommending system based on social media according to claim 7, it is characterised in that the recommended by client
Module running is comprised the following steps：
3-1) dependency by user and commodity obtains the items list that may recommend；
The analysis of the dependency of user and commodity 3-2) is carried out to the good friend of user, and by the items list of good friend to the user
Voted；
3-3) commercial product recommending is further processed by analyzing user's portrait, target, user's portrait bag are recommended in subdivision
Include age, income and interest；
Commercial product recommending is carried out for the user by the interactive mode of social media 3-4), the interactive mode of the social media includes
Add good friend, quote good friend, personal letter and comment.
9. the commercial product recommending system based on social media according to claim 7, it is characterised in that the commercial product recommending mould
Block running is comprised the following steps：
4-1) dependency by commodity and user obtains user that may be interested in the commodity；
The correlation analysiss of commodity and user are carried out to the good friend of user 4-2), and the user is entered by the items list of good friend
Row ballot；
4-3) commercial product recommending is further processed by analyzing user's portrait, target, user's portrait bag are recommended in subdivision
Include age, income and interest；
Commercial product recommending is carried out for the user by the interactive mode of social media 4-4), the interactive mode of the social media includes
Add good friend, quote good friend, personal letter and comment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410110393.1A CN103886074B (en) | 2014-03-24 | 2014-03-24 | Commercial product recommending system based on social media |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410110393.1A CN103886074B (en) | 2014-03-24 | 2014-03-24 | Commercial product recommending system based on social media |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103886074A CN103886074A (en) | 2014-06-25 |
| CN103886074B true CN103886074B (en) | 2017-03-15 |
Family
ID=50954966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410110393.1A Active CN103886074B (en) | 2014-03-24 | 2014-03-24 | Commercial product recommending system based on social media |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103886074B (en) |
Families Citing this family (72)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104636392B (en) | 2013-11-13 | 2018-07-27 | 腾讯科技（北京）有限公司 | Carry out method, system, server and browser that recommendation information issues |
| CN105893361A (en) * | 2014-09-23 | 2016-08-24 | 江苏奥博洋信息技术有限公司 | A method for structured classification of users in socialized media |
| CN105528704A (en) * | 2014-09-28 | 2016-04-27 | 南京理工大学常熟研究院有限公司 | Dynamic product recommendation method based on stage |
| CN104317920A (en) * | 2014-10-29 | 2015-01-28 | 中英融贯资讯（武汉）有限公司 | Information pushing method and information pushing system based on intelligent user demand matching |
| CN104731852A (en) * | 2014-12-16 | 2015-06-24 | 芜湖乐锐思信息咨询有限公司 | Big data system |
| KR102307786B1 (en) * | 2014-12-19 | 2021-10-05 | 삼성전자주식회사 | Electronic apparatus and user profile estimating method thereof |
| US20160180723A1 (en) * | 2014-12-22 | 2016-06-23 | Intel Corporation | Context derived behavior modeling and feedback |
| CN105869001A (en) * | 2015-01-19 | 2016-08-17 | 苏宁云商集团股份有限公司 | Customized product recommendation guiding method and system |
| CN104616178A (en) * | 2015-03-06 | 2015-05-13 | 浪潮集团有限公司 | Recommendation method of E-commerce goods based on big-data multi-label classification method |
| EP3076359A1 (en) | 2015-04-01 | 2016-10-05 | Tata Consultancy Services Ltd. | Implementing retail customer analytics data model in a distributed computing environment |
| CN105809467A (en) * | 2015-04-28 | 2016-07-27 | 成都鹏业软件股份有限公司 | Instant communication tool-based commodity recommending system |
| CN106202078A (en) * | 2015-04-30 | 2016-12-07 | 雅虎公司 | Object hunting system and method and use the mobile communications device of this object method for searching |
| CN104933239A (en) * | 2015-06-09 | 2015-09-23 | 江苏大学 | Hybrid model based personalized position information recommendation system and realization method therefor |
| CN106294515B (en) * | 2015-06-11 | 2020-08-21 | 腾讯科技（深圳）有限公司 | Promotion information processing method and device |
| CN104992060A (en) * | 2015-06-25 | 2015-10-21 | 腾讯科技（深圳）有限公司 | User age estimation method and apparatus |
| CN105046274A (en) * | 2015-07-13 | 2015-11-11 | 浪潮软件集团有限公司 | Automatic labeling method for electronic commerce commodity category |
| CN105138551A (en) * | 2015-07-14 | 2015-12-09 | 青岛海信传媒网络技术有限公司 | Method and apparatus for obtaining user interest tag |
| CN106485525A (en) * | 2015-08-31 | 2017-03-08 | 百度在线网络技术（北京）有限公司 | Information processing method and device |
| CN105282702B (en) * | 2015-09-07 | 2018-01-19 | 广东欧珀移动通信有限公司 | A kind of indoor localization method and user terminal |
| CN105222771B (en) * | 2015-09-07 | 2018-06-29 | 广东欧珀移动通信有限公司 | A kind of indoor article localization method and user terminal |
| CN105117475B (en) * | 2015-09-07 | 2018-01-26 | 广东欧珀移动通信有限公司 | Recommend method and user terminal in a kind of shop |
| CN106649367B (en) * | 2015-10-30 | 2020-03-03 | 北京国双科技有限公司 | Method and device for detecting keyword popularization degree |
| CN106919585A (en) * | 2015-12-24 | 2017-07-04 | 中移（杭州）信息技术有限公司 | URL according to terminal determines the method and device of merchandise news |
| CN106970914B (en) * | 2016-01-12 | 2020-12-22 | 北京京东尚科信息技术有限公司 | Friend recommendation method and device based on social e-commerce |
| US10726465B2 (en) | 2016-03-24 | 2020-07-28 | International Business Machines Corporation | System, method and computer program product providing eye tracking based cognitive filtering and product recommendations |
| CN105704014B (en) * | 2016-03-28 | 2019-05-28 | 跨理（广州）供应链管理有限公司 | The method of integration shopping and social activity, the system and shopping network platform for being integrated with shopping and social activity |
| CN106055618B (en) * | 2016-05-26 | 2020-02-07 | 优品财富管理有限公司 | Data processing method based on web crawler and structured storage |
| CN106168971A (en) * | 2016-07-08 | 2016-11-30 | 北京麒麟合盛网络技术有限公司 | information subscribing method and device |
| CN107633416B (en) * | 2016-07-18 | 2021-07-09 | 阿里巴巴集团控股有限公司 | Method, device and system for recommending service object |
| CN106294846A (en) * | 2016-08-19 | 2017-01-04 | 维沃移动通信有限公司 | A kind of recommendation method of information and mobile terminal |
| CN106296312A (en) * | 2016-08-30 | 2017-01-04 | 江苏名通信息科技有限公司 | Online education resource recommendation system based on social media |
| CN106411846A (en) * | 2016-08-30 | 2017-02-15 | 江苏名通信息科技有限公司 | Connecting system and method for network game and e-commerce |
| CN106503096B (en) * | 2016-10-14 | 2020-02-04 | 上海斐讯数据通信技术有限公司 | Social network recommendation method and system based on distributed noise interference prevention |
| CN106503863A (en) * | 2016-11-10 | 2017-03-15 | 北京红马传媒文化发展有限公司 | Based on the Forecasting Methodology of the age characteristicss of decision-tree model, system and terminal |
| CN106779827A (en) * | 2016-12-02 | 2017-05-31 | 上海晶樵网络信息技术有限公司 | A kind of Internet user's behavior collection and the big data method of analysis detection |
| CN106874093B (en) * | 2017-02-14 | 2021-09-14 | 阿里巴巴（中国）有限公司 | Method for calculating target crowd based on user portrait, calculation engine and calculation equipment |
| CN106875280A (en) * | 2017-03-02 | 2017-06-20 | 深圳大图科创技术开发有限公司 | Integrated community service platform |
| CN106874891A (en) * | 2017-03-16 | 2017-06-20 | 湖南众益文化传媒股份有限公司 | Smart media ad system based on recognition of face |
| CN107220831A (en) * | 2017-04-06 | 2017-09-29 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | A kind of user group's division methods and system based on anti-fake traceability system |
| CN106960063A (en) * | 2017-04-20 | 2017-07-18 | 广州优亚信息技术有限公司 | A kind of internet information crawl and commending system for field of inviting outside investment |
| CN107103499A (en) * | 2017-04-26 | 2017-08-29 | 中国人民大学 | Method is recommended in a kind of cross-site cold start-up based on deep learning |
| CN108932646B (en) * | 2017-05-26 | 2021-04-13 | 中移（杭州）信息技术有限公司 | User tag verification method and device based on operator and electronic equipment |
| CN108053261B (en) * | 2017-12-27 | 2021-05-14 | 网智天元科技集团股份有限公司 | Marketing information processing method and system |
| CN108399561A (en) * | 2018-02-10 | 2018-08-14 | 梧州井儿铺贸易有限公司 | A kind of commercial product recommending system that recommendation effect is good |
| US11514333B2 (en) * | 2018-04-30 | 2022-11-29 | Meta Platforms, Inc. | Combining machine-learning and social data to generate personalized recommendations |
| CN108830677A (en) * | 2018-05-28 | 2018-11-16 | 苏州若依玫信息技术有限公司 | A kind of e-commerce system based on data analysis |
| CN108920537A (en) * | 2018-06-11 | 2018-11-30 | 颜彦 | The internet proposed algorithm of media service platform is melted in a kind of application |
| CN110070410A (en) * | 2019-03-07 | 2019-07-30 | 特斯联（北京）科技有限公司 | A kind of population social activity analysis method and system based on big data |
| CN110188288B (en) * | 2019-05-29 | 2021-05-07 | 山东大学 | Information recommendation method, system, medium and device based on multiple modes |
| CN110197404B (en) * | 2019-06-12 | 2022-04-01 | 合肥工业大学 | Personalized long-tail commodity recommendation method and system capable of reducing popularity deviation |
| CN110489645A (en) * | 2019-08-15 | 2019-11-22 | 深圳市云积分科技有限公司 | A kind of brand association consumer orients the information processing method and device of marketing |
| CN110503467B (en) * | 2019-08-15 | 2022-11-22 | 深圳市云积分科技有限公司 | Cross-platform consumer group acquisition method and device |
| CN110827119A (en) * | 2019-10-18 | 2020-02-21 | 用友网络科技股份有限公司 | Information interaction system based on WeChat applet |
| CN110968795B (en) * | 2019-11-27 | 2023-06-02 | 国网能源研究院有限公司 | Data association matching system of company image lifting system |
| CN111028060B (en) * | 2019-12-11 | 2023-08-29 | 上海风秩科技有限公司 | Commodity recommendation method, commodity recommendation device, computer equipment and readable storage medium |
| CN111177581A (en) * | 2019-12-25 | 2020-05-19 | 清华大学 | Multi-platform-based social e-commerce website commodity recommendation method and device |
| CN111507804B (en) * | 2020-04-21 | 2022-05-13 | 莫毓昌 | Emotion perception commodity recommendation method based on mixed information fusion |
| CN111611484B (en) * | 2020-05-13 | 2023-08-11 | 湖南微步信息科技有限责任公司 | Stock recommendation method and system based on article attribute identification |
| CN111833096A (en) * | 2020-06-10 | 2020-10-27 | 北京龙云科技有限公司 | Advertisement recommendation method and system based on machine learning |
| WO2022021383A1 (en) * | 2020-07-31 | 2022-02-03 | 深圳齐心集团股份有限公司 | Stationery recommendation system |
| CN112085568B (en) * | 2020-09-09 | 2023-10-27 | 百度在线网络技术（北京）有限公司 | Commodity and rich media aggregation display method and equipment, electronic equipment and medium |
| CN112348549A (en) * | 2020-09-16 | 2021-02-09 | 广州阿凡提电子科技有限公司 | Advertisement operation method, intelligent terminal and storage device |
| CN111882409B (en) * | 2020-09-28 | 2020-12-08 | 武汉卓尔数字传媒科技有限公司 | Method and device for recommending main body and electronic equipment |
| CN112269504B (en) * | 2020-10-22 | 2022-09-02 | 维沃移动通信有限公司 | Information display method and device and electronic equipment |
| CN112435067A (en) * | 2020-11-30 | 2021-03-02 | 翼果（深圳）科技有限公司 | Intelligent advertisement putting method and system for cross-e-commerce platform and social platform |
| TWI807243B (en) * | 2021-01-12 | 2023-07-01 | 統一超商股份有限公司 | An online social system and method with crowdfunding |
| CN112800298B (en) * | 2021-02-01 | 2021-11-12 | 黄秋奋 | Internet-based electronic commerce data processing method and system |
| CN112801761A (en) * | 2021-04-13 | 2021-05-14 | 中智关爱通(南京)信息科技有限公司 | Commodity recommendation method, computing device and computer-readable storage medium |
| CN113837824A (en) * | 2021-08-16 | 2021-12-24 | 上海漫集信息科技有限公司 | Information pushing method and system |
| CN113505891A (en) * | 2021-09-09 | 2021-10-15 | 北京轻松筹信息技术有限公司 | Age information determining method and device, and information pushing method and device |
| CN116089740B (en) * | 2023-03-11 | 2023-11-07 | 深圳市针线科技有限公司 | Group purchase information retrieval method and system based on social network |
| CN117708437B (en) * | 2024-02-05 | 2024-04-16 | 四川日报网络传媒发展有限公司 | Recommendation method and device for personalized content, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102317941A (en) * | 2011-07-30 | 2012-01-11 | 华为技术有限公司 | Information recommending method, recommending engine and network system |
| CN102779312A (en) * | 2012-07-23 | 2012-11-14 | 魏珂 | Method and system for recommending and delivering gift based on social network |
| CN102903047A (en) * | 2011-07-26 | 2013-01-30 | 阿里巴巴集团控股有限公司 | Method and device for releasing commodity information |
| CN103177107A (en) * | 2013-03-27 | 2013-06-26 | 百度在线网络技术（北京）有限公司 | Recommendation method and device |
| CN103544607A (en) * | 2012-07-11 | 2014-01-29 | 北京长生天地电子商务有限公司 | Information processing system and information processing method for implementing network transaction by aid of social network |
-
2014
- 2014-03-24 CN CN201410110393.1A patent/CN103886074B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102903047A (en) * | 2011-07-26 | 2013-01-30 | 阿里巴巴集团控股有限公司 | Method and device for releasing commodity information |
| CN102317941A (en) * | 2011-07-30 | 2012-01-11 | 华为技术有限公司 | Information recommending method, recommending engine and network system |
| CN103544607A (en) * | 2012-07-11 | 2014-01-29 | 北京长生天地电子商务有限公司 | Information processing system and information processing method for implementing network transaction by aid of social network |
| CN102779312A (en) * | 2012-07-23 | 2012-11-14 | 魏珂 | Method and system for recommending and delivering gift based on social network |
| CN103177107A (en) * | 2013-03-27 | 2013-06-26 | 百度在线网络技术（北京）有限公司 | Recommendation method and device |
Non-Patent Citations (1)
| Title |
|---|
| 基于社会网络的主动信息推送算法研究;余天豪;《中国优秀硕士学位论文全文数据库》;20121215(第2012年12期);正文第8页第2.2.1节第1-3段、第23页第3.4.2节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103886074A (en) | 2014-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103886074B (en) | Commercial product recommending system based on social media | |
| US11507551B2 (en) | Analytics based on scalable hierarchical categorization of web content | |
| US11301761B2 (en) | Behavioral prediction for targeted end users | |
| US20190294642A1 (en) | Website fingerprinting | |
| US8886636B2 (en) | Context transfer in search advertising | |
| CN106296312A (en) | Online education resource recommendation system based on social media | |
| CN110222272A (en) | A kind of potential customers excavate and recommended method | |
| US11238233B2 (en) | Artificial intelligence engine for generating semantic directions for websites for automated entity targeting to mapped identities | |
| Lu et al. | Is user-generated content always helpful? The effects of online forum browsing on consumers' travel purchase decisions | |
| CN107665444A (en) | A kind of web advertisement acute effect evaluation method and system based on the online behavior of user | |
| CN102629257A (en) | Commodity recommending method of e-commerce website based on keywords | |
| CN104254851A (en) | Method and system for recommending content to a user | |
| Siddiqui et al. | Web mining techniques in e-commerce applications | |
| CN104252679A (en) | Construction method and system of brand advertisement evaluation system | |
| JP2019125007A (en) | Information analyzer, information analysis method and information analysis program | |
| CN105023178B (en) | A kind of electronic commerce recommending method based on ontology | |
| US20110313842A1 (en) | System and methods thereof for providing an advertisement placement recommendation based on trends | |
| Alazab et al. | Maximising competitive advantage on E-business websites: A data mining approach | |
| Mei et al. | Overview of Web mining technology and its application in e-commerce | |
| Dias et al. | Automating the extraction of static content and dynamic behaviour from e-commerce websites | |
| Henrys | Importance of web scraping in e-commerce and e-marketing | |
| Han et al. | Fitnet: Identifying fashion influencers on twitter | |
| CN110717089A (en) | User behavior analysis system and method based on weblog | |
| KR20100046421A (en) | Method and server for estimating preference of commodity | |
| Guan et al. | Predicting time-bounded purchases during a mega shopping festival |
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 | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20180411 Address after: 210019 room 1201, 12 building, 31 Guang Jie Road, Jianye District, Nanjing, Jiangsu, China Patentee after: NANJING JIANGDA BODA INFORMATION TECHNOLOGY CO.,LTD. Address before: Zhenjiang City, Jiangsu Province, 212000 Jingkou District Road No. 118 Jingkou high-tech business center 6 floor Patentee before: JIANGSU MINGTONG TECHNOLOGY CO.,LTD. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240308 Address after: No. 108 Qingchun Road, Xinminzhou, Jingkou District, Zhenjiang City, Jiangsu Province, 212000 Patentee after: Jiangsu Zhenqu Unlimited Information Technology Co.,Ltd. Country or region after: China Address before: Room 1201, 12th Floor, No. 31 Guangju Road, Jianye District, Nanjing City, Jiangsu Province, 210019 Patentee before: NANJING JIANGDA BODA INFORMATION TECHNOLOGY CO.,LTD. Country or region before: China |
|
| TR01 | Transfer of patent right |