CN105608111A - Method and system for recommending object to terminal user - Google Patents
Method and system for recommending object to terminal user Download PDFInfo
- Publication number
- CN105608111A CN105608111A CN201510912090.6A CN201510912090A CN105608111A CN 105608111 A CN105608111 A CN 105608111A CN 201510912090 A CN201510912090 A CN 201510912090A CN 105608111 A CN105608111 A CN 105608111A
- Authority
- CN
- China
- Prior art keywords
- associated group
- recommendation degree
- group
- association
- frequency
- 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
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
Abstract
The invention provides a method for recommending an object to a terminal user. The method comprises the steps of based on N objects, establishing an object correlation data set, wherein the object correlation data set comprises all objects in the N objects browsed by any terminal user within a first time period, and a correlation group consisting of two objects in the N objects successively browsed by any terminal user within the first time period according to a browsing sequence; determining the individual frequency of each object in the object correlation data set, occurring in the object correlation data set and the correlation group frequency of the object correlation group occurring in the object correlation data set, calculating the correlation degree of the two objects in the object correlation group based on the individual frequency of each object in the object correlation group and the correlation group frequency, and determining the object calibration recommendation degree for a terminal user in the objects based on the correlation degree and the determined seed object.
Description
Technical field
The present invention relates to network data processing, relate more particularly to the technology of terminal use's recommended.
Background technology
Commending system is that the industries such as internet industry, ecommerce and retail business are the most typical and at Data MiningImportant application.
At present, the commending system of most of company is all realized based on collaborative filtering, although collaborative filtering is being answeredDemonstrate good performance when with commercial product recommending, but also had improvable place. Collaborative filtering is based between userConsumption similarity build the evaluation to user, thereby make recommendation, and similarity between user is sparse by higher-dimensionSimilarity between vector identifies, and the similarity between the user that such evaluation method obtains may be inaccurate, thereby cause the inaccuracy of recommendation results, and along with the increasing of trade company's quantity, the vector that commending system calculates is rareThin degree can increase, and this has not only reduced the degree of accuracy of recommending, increased the complexity of calculating simultaneously, has reduced commending systemPerformance.
Summary of the invention
A kind of method to terminal use's recommended is provided herein, and it comprises:
A. based on N object, set up object association data collection, this object association data collection comprises in very first time section by anyAll objects in described N the object that terminal use browsed, and successively browsed by any terminal use in very first time sectionTwo objects in the N a crossing object are by the associated group of browsing order composition;
B. determine that the concentrated each object of described object association data is in the concentrated frequency of individuals occurring of this object association data;
C. determine that described object association group is in the concentrated associated group frequency occurring of this object association data;
D. the frequency of individuals based on each object in described object association group and this associated group frequency computation part described in this object closeThe degree of association of two objects in joint group;
E. determine the object that is had the terminal use of recommended requirements to browse in a described N object, set it as seed object;
F. demarcate the recommendation degree of described seed object;
G. concentrate and search the first kind associated group that comprises described seed object in described object association data, and be described firstThe object of not yet being demarcated in class associated group is demarcated recommendation degree; And
H. concentrate to search in described object association data and comprise that the Equations of The Second Kind of the non-seed object in described first kind associated group closesJoint group, and be the object demarcation recommendation degree of not demarcating recommendation degree in this Equations of The Second Kind associated group.
According to an aspect of the present invention, this is in the method for terminal use's recommended, and described steps d is according to following public affairsFormula is determined the described degree of association:
WhereinFor the frequency of individuals of object x,For the frequency of individuals of object y,For the associated group xy of object x and y compositionClass frequency, Wxy is the degree of association of object x and y in associated group xy.
According to an aspect of the present invention, this in the method for terminal use's recommended, the recommendation of described seed objectDegree is 1, and described recommendation degree is determined according to following formula:
Wherein, x is the object of having demarcated recommendation degree in associated group xy, and its recommendation degree is Rx, and y does not demarcate in associated group xyThe object of recommendation degree, Ry represents the recommendation degree of y, p is the recommendation degree probability of spreading of being scheduled to.
According to example of the present invention, also provide to the system of terminal use's recommended, it comprises:
First module, it,, for based on N object, sets up object association data collection, and this object association data collection comprises at first o'clockBetween in section by all objects in described N the object that arbitrarily terminal use browsed, and in very first time end by whole arbitrarilyTwo objects in N the object that end subscriber was successively browsed are by the associated group of browsing order composition;
Second unit, it concentrates each object to concentrate in this object association data occur individual for determining described object association dataBody frequency;
Unit the 3rd, it concentrates for determining described object association group the associated group frequency occurring in this object association data;
Unit the 4th, it is for the frequency of individuals based on the each object of described object association group and this associated group frequency computation partThe degree of association between object in object association group described in this;
Unit the 5th, its object for determining that a described N object is had the terminal use of recommended requirements to browse, by its workFor seed object;
Unit the 6th, it is for demarcating the recommendation degree of described seed object;
Unit the 7th, it is for concentrating and search the first kind associated group that comprises described seed object in described object association data,And be the non-seed object demarcation recommendation degree in described first kind associated group; And
Unit the 8th, it is for concentrating and search the non-seed pair that comprises described first kind associated group in described object association dataThe Equations of The Second Kind associated group of elephant, and be the object demarcation recommendation degree of not demarcating recommendation degree in this Equations of The Second Kind associated group.
According to an aspect of the present invention, provide in the system of terminal use's recommended, described step the fourThe described degree of association is determined according to following formula in unit:
WhereinFor the frequency of individuals of object x,For the frequency of individuals of object y,For the associated group xy of object x and y compositionClass frequency, Wxy is the degree of association of object x and y in associated group xy.
According to an aspect of the present invention, provide in the system of terminal use's recommended, described seed objectRecommendation degree be 1, described recommendation degree according to following formula determine:
Wherein, x is the object of having demarcated recommendation degree in associated group xy, and its recommendation degree is Rx, and y does not demarcate in associated group xyThe object of recommendation degree, Ry represents the recommendation degree of y, p is the recommendation degree probability of spreading of being scheduled to.
Brief description of the drawings
Fig. 1 is the flow chart of the method to terminal use's recommended of an example according to the present invention.
Fig. 2 is the structural representation of the system to terminal use's recommended of an example according to the present invention.
Detailed description of the invention
Referring now to accompanying drawing, schematic example of the present invention is described. Identical drawing reference numeral represents identical element. BelowThe each embodiment describing contributes to those skilled in the art thoroughly to understand the present invention, and is intended to example and unrestricted. Unless separately hadLimit, the term (comprising science, technology and industry slang) using in literary composition has with those skilled in the art in the invention generalAll over the identical implication of implication of understanding. In addition, in flow chart, the sequencing of each step is not also limited with illustrated order.
Fig. 1 is the flow chart of the method to terminal use's recommended of an example according to the present invention. As shown in the figure, existStep 10, based on N object, sets up object association data collection, and this object association data collection comprises in very first time section by anyAll objects in described N the object that terminal use browsed, and successively browsed by any terminal use in very first time sectionTwo objects in the N a crossing object are by the associated group of browsing order composition.
Typically, N is a larger data, and very first time section is also the time period of waiting such as one month, thus, and instituteThe object association data collection obtaining is a data set that data content is large. At this, for ease of setting forth, suppose that N is 4, that is, supposeHave A, B, C, tetra-objects of D, very first time section is 1 day. In the time of this day, party a subscriber has first been browsed A, then has browsedC, and then browsed D, later browse again again A; User's second has first been browsed A, then has browsed D, and then goes back to have browsedA, then remove to have browsed B; The third user has browsed B, has browsed afterwards C; Fourth user browses order for D, B, C, A. Thus, this exampleIn, all objects in this N object of being browsed by first, second, third, fourth in this very first time section be A, C, D, A, A, D, A, B,B, C, D, B, C, A; Two objects in N the object of successively being browsed by party a subscriber in very first time section are by browsing order compositionAssociated group comprise AC, CD, DA, two objects in N the object that user's second was successively browsed are by the pass of browsing order and formingJoint group comprises AD, DA, AB, and two objects in N the object that user third successively browsed are by the associated group of browsing order compositionComprise BC, two objects in N the object that user's fourth was successively browsed by browse associated group that order forms comprise DB, BC,CA. And then, in this example, based on these 4 objects set up object association data collection comprise A, C, D, A, A, D, A, B, B, C, D, B,C、A、AC、CD、DA、AD、DA、AB、BC、DB、BC、CA。
Visible, build in the process of this associated data set, if certain object by same terminal use in the different timeBrowsed repeatedly, this object should be concentrated and occur repeatedly in this associated data, and in this associated group, in associated group, each group isBrowse the tactic of two objects in group according to user, and, if same terminal use is clear at different time order and functionsThe situation of two objects of looking at occurred repeatedly, and this associated group should be concentrated and occur repeatedly in this associated data. Such as, above thisConcrete example, in this morning of day, user third successively browsed B and C, thereby formed these two objects by browsing der groupBecome associated group BC, and user browsed B and C the third afternoon again, again formed associated group BC, correspondingly, given above rightResemble associated data set will be revised as comprise A, C, D, A, A, D, A, B, B, C, D, B, C, A, AC, CD, DA, AD, DA, AB, BC, BC,DB、BC、CA。
In step 12, determine that the concentrated each object of described object association data is at the concentrated individuality occurring of this object association dataFrequency. It is the frequency of individuals of A that the object association data that object A sets up in step 10 is concentrated the number of times occurring, object BThe number of times occurring becomes the frequency of individuals of B, the like.
In step 14, determine that described object association group is in the concentrated associated group frequency occurring of this object association data. ObjectThe frequency of associated group refers to the associated group that comprises two objects with same sequence and concentrates and occur in whole object association dataNumber of times. For example, the associated group frequency representation of associated group AB is。
In step 16, the frequency of individuals based on each object in described object association group and this associated group frequency computation part shouldThe degree of association in described object association group between object. If the degree of association of object A and object B in compute associations group AB, rootAccording to the frequency of individuals of object A, object B frequency of individuals, and the associated group frequency of associated group ABCome according to as followsEquation (1) calculate:
（1）
Wherein,For the frequency of individuals of object x,For the frequency of individuals of object y,For the associated group xy of object x and y compositionClass frequency, Wxy is the degree of association of object x and y in associated group xy.
If object x is object A, and object y is object B, according to equation (1): the calculation of relationship degree of A and B is as follows:
In step 18, determine the object that is had the terminal use of recommended requirements to browse in a described N object, set it as seedObject. In conjunction with example given above, suppose to have recommended requirements terminal use be the third, first from N object, determineGo out the third object of once browsing, as described in step 10, third once browsed B and C, therefore, and for there being recommendation to needAsk terminal use's the third object B and C be all confirmed as seed object.
In step 20, demarcate the recommendation degree of described seed object. According to an example of the present invention, can be by seed objectRecommend scale to be decided to be 1.
In step 22, concentrate and search the first kind associated group that comprises this seed object in described object association data, and beThe object of not yet being demarcated in this first kind associated group of searching is demarcated recommendation degree.
According to example of the present invention, if the object in the associated group finding demarcated, without further markFixed, the object that only need to not demarcate those is demarcated, wherein, and each pass in the one or more associated group that findIn joint group, not demarcating the recommendation degree of object determines according to following equation (2):
（2）
Wherein, x represents the object of having been demarcated recommendation degree in associated group xy, and its recommendation degree is Rx, and y is in associated group xyThe object of not demarcating recommendation degree, Ry represents the recommendation degree of y, and p is the recommendation degree probability of spreading of being scheduled to, and Wxy is right in associated group xyResemble the degree of association of x and y.
Still with reference to example given above, in step 22, first find out the associated group that comprises B, have respectively AB, BC,DB. Wherein, C is also seed object, and its recommendation degree will be demarcated as 1, thus, determines the recommendation degree of A and D according to equation (2):
Wherein,Value be 1,According to equation (1) calculate obtain, thus, can by p,、And minimum in 1Value is defined as。
In step 24, concentrate and search the non-seed object comprising in described first kind associated group in described object association dataEquations of The Second Kind associated group, and be the object demarcation recommendation degree of not demarcating recommendations degree in this Equations of The Second Kind associated group. As true in step 22The fixed associated group that comprises B, has respectively AB, BC, DB, can determine that the non-seed object relating to comprises A and D, as above basisEquation (2) has been determined the recommendation degree of A and D. First find out the associated group that comprises A and D, determining each associated group according to equation (2)In do not demarcated the recommendation degree of the object of recommendation degree, if two objects in the associated group finding have all been demarcated recommendationDegree, no longer computes repeatedly. For example find associated AB and BA, DB, DC or CD etc. can skip over.
Perform step 24, until do not have more object to be demarcated always.
According to a concrete example of the present invention, object can be trade company, and terminal use is consumption user, thus,The object that terminal use according to the present invention browsed for example refers to the trade company of consumption user post-consumer, and this consumption data can be fromThe transaction record of businessman obtains. And alternatively, the object that terminal use browsed also can refer to the electric business that user browsedOn-line shop, browse record can by trade company site browse record and terminal use's IP or the pet name etc. be obtained.
According to the present invention, the system to terminal use's recommended of each example can realize by software, but also can be by hardPart is realized, or realizes by the combination of software and hardware. Adopt according to the side to terminal use's recommended shown in the presentMethod, by independent cascade model, has increased the effective candidate trade company number of trade company's correlation recommendation, has improved the precision of recommending.
The present invention also provides to the system of terminal use's recommended, and Fig. 2 is the structural representation of this system. As scheme instituteShow, this system to terminal use's recommended comprise first module 40, second unit 42, the 3rd unit 44, the 4th unit 46,The 5th unit 48, the 6th unit 50, the 7th unit 52 and the 8th unit 54.
First module 40, for based on N object, is set up object association data collection, and this object association data collection comprises firstAll objects in time period in quilt described N the object that terminal use browsed arbitrarily, and very first time end is interior by anyTwo objects in N the object that terminal use successively browsed are by the associated group of browsing order composition. Wherein, setting up object closesThe process of connection data set is basically identical with above integrating step 10 is introduced, repeats no more.
Second unit 42 concentrates each object to concentrate and occur in this object association data for determining described object association dataFrequency of individuals. The associated group frequency occurring is concentrated for determining described object association group in the 3rd unit 44 in this object association dataRate. The 4th unit 46 should for the frequency of individuals based on the each object of described object association group and this associated group frequency computation partThe degree of association in described object association group between object. Determining of frequency of individuals, associated group frequency and associated group frequency is as above civilianIntegrating step 12,14 and 16 introduce like that, and in associated group between object the degree of association really normal root carry out according to equation (1).
The object of the 5th unit 48 for determining that a described N object is had the terminal use of recommended requirements to browse, willIt is as seed object. The 6th unit 50 is for demarcating the recommendation degree of described seed object. As 18 and 20 of integrating steps aboveIntroduce, the recommendation scale of seed object is decided to be 1.
The 7th unit 52 is for concentrating and search the first kind association that comprises described seed object in described object association dataGroup, and be the non-seed object demarcation recommendation degree in described first kind associated group. The 8th unit 54 is at described object associationData centralization is searched the Equations of The Second Kind associated group that comprises the non-seed object in described first kind associated group, and is this Equations of The Second Kind associationThe object of not demarcating recommendation degree in group is demarcated recommendation degree. First kind associated group can be with reference to being above combined with Equations of The Second Kind associated groupStep 22 and 24 is introduced, and the determining of recommendation degree of the object of demarcation recommendation degree can not determined by equation (2).
According to the present invention, the system to terminal use's recommended of example can realize by software, also can be real by hardwareExisting, or realize by the combination of software and hardware.
Although, in conjunction with having described the present invention above, should be understood that the each example in literary composition can mutually combine. Do not deviating from thisIn the scope of disclosure of the invention and the situation of spirit, also should fall into the application institute to the amendment of each step or parts, unit in exampleIn the scope of attached claims.
Claims (6)
1. to a method for terminal use's recommended, it comprises:
A. based on N object, set up object association data collection, this object association data collection comprises in very first time section by anyAll objects in described N the object that terminal use browsed, and successively clear by any terminal use in very first time sectionTwo objects in the N a looking at object are by the associated group of browsing order composition;
B. determine that the concentrated each object of described object association data is in the concentrated frequency of individuals occurring of this object association data;
C. determine that described object association group is in the concentrated associated group frequency occurring of this object association data;
D. the frequency of individuals based on each object in described object association group and this associated group frequency determine that object closes described in thisThe degree of association of two objects in joint group;
E. determine the object that is had the terminal use of recommended requirements to browse in a described N object, set it as seed object;
F. demarcate the recommendation degree of described seed object;
G. concentrate and search the first kind associated group that comprises described seed object in described object association data, and be described firstThe object of not yet being demarcated in class associated group is demarcated recommendation degree; And
H. concentrate to search in described object association data and comprise that the Equations of The Second Kind of the non-seed object in described first kind associated group closesJoint group, and be the object demarcation recommendation degree of not demarcating recommendation degree in this Equations of The Second Kind associated group.
2. the method for claim 1, wherein described steps d is determined the described degree of association according to following formula:
WhereinFor the frequency of individuals of object x,For the frequency of individuals of object y,For the associated group xy of object x and y compositionClass frequency, Wxy is the degree of association of object x and y in associated group xy.
3. method as claimed in claim 2, wherein, the recommendation degree of described seed object is 1, described recommendation degree is according to following public affairsFormula is determined:
Wherein, x is the object of having demarcated recommendation degree in associated group xy, and its recommendation degree is Rx, and y does not demarcate in associated group xyThe object of recommendation degree, Ry represents the recommendation degree of y, p is the recommendation degree probability of spreading of being scheduled to.
4. to a system for terminal use's recommended, it comprises:
First module, it,, for based on N object, sets up object association data collection, and this object association data collection comprises at first o'clockBetween in section by all objects in described N the object that arbitrarily terminal use browsed, and in very first time end by whole arbitrarilyTwo objects in N the object that end subscriber was successively browsed are by the associated group of browsing order composition;
Second unit, it concentrates each object to concentrate in this object association data occur individual for determining described object association dataBody frequency;
Unit the 3rd, it concentrates for determining described object association group the associated group frequency occurring in this object association data;
Unit the 4th, it is for the frequency of individuals based on the each object of described object association group and this associated group frequency computation partThe degree of association between object in object association group described in this;
Unit the 5th, its object for determining that a described N object is had the terminal use of recommended requirements to browse, by its workFor seed object;
Unit the 6th, it is for demarcating the recommendation degree of described seed object;
Unit the 7th, it is for concentrating and search the first kind associated group that comprises described seed object in described object association data,And be the non-seed object demarcation recommendation degree in described first kind associated group; And
Unit the 8th, it is for concentrating and search the non-seed pair that comprises described first kind associated group in described object association dataThe Equations of The Second Kind associated group of elephant, and be the object demarcation recommendation degree of not demarcating recommendation degree in this Equations of The Second Kind associated group.
5. system as claimed in claim 4, wherein, the described degree of association is determined according to following formula in described step the Unit four:
WhereinFor the frequency of individuals of object x,For the frequency of individuals of object y,For the associated group xy of object x and y compositionClass frequency, Wxy is the degree of association of object x and y in associated group xy.
6. system as claimed in claim 5, wherein, the recommendation degree of described seed object is 1, described recommendation degree is according to following public affairsFormula is determined:
Wherein, x is the object of having demarcated recommendation degree in associated group xy, and its recommendation degree is Rx, and y does not demarcate in associated group xyThe object of recommendation degree, Ry represents the recommendation degree of y, p is the recommendation degree probability of spreading of being scheduled to.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510912090.6A CN105608111B (en) | 2015-12-11 | 2015-12-11 | To the method and system of terminal user's recommended |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510912090.6A CN105608111B (en) | 2015-12-11 | 2015-12-11 | To the method and system of terminal user's recommended |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105608111A true CN105608111A (en) | 2016-05-25 |
| CN105608111B CN105608111B (en) | 2019-02-15 |
Family
ID=55988051
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510912090.6A Active CN105608111B (en) | 2015-12-11 | 2015-12-11 | To the method and system of terminal user's recommended |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105608111B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107885774A (en) * | 2017-09-29 | 2018-04-06 | 北京京东尚科信息技术有限公司 | Data processing method and system |
| CN109460519A (en) * | 2018-12-28 | 2019-03-12 | 上海晶赞融宣科技有限公司 | Browse object recommendation method and device, storage medium, server |
| CN111507768A (en) * | 2020-04-17 | 2020-08-07 | 腾讯科技（深圳）有限公司 | Determination method of potential user, model training method and related device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521420A (en) * | 2012-01-04 | 2012-06-27 | 西安电子科技大学 | Socialized filtering method on basis of preference model |
| CN102789462A (en) * | 2011-05-18 | 2012-11-21 | 阿里巴巴集团控股有限公司 | Project recommendation method and system |
| CN103177093A (en) * | 2013-03-13 | 2013-06-26 | 北京开心人信息技术有限公司 | General recommendation method and system based on object tags |
| US20140236879A1 (en) * | 2009-07-08 | 2014-08-21 | Ebay Inc. | Systems and methods for contextual recommendations |
-
2015
- 2015-12-11 CN CN201510912090.6A patent/CN105608111B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140236879A1 (en) * | 2009-07-08 | 2014-08-21 | Ebay Inc. | Systems and methods for contextual recommendations |
| CN102789462A (en) * | 2011-05-18 | 2012-11-21 | 阿里巴巴集团控股有限公司 | Project recommendation method and system |
| CN102521420A (en) * | 2012-01-04 | 2012-06-27 | 西安电子科技大学 | Socialized filtering method on basis of preference model |
| CN103177093A (en) * | 2013-03-13 | 2013-06-26 | 北京开心人信息技术有限公司 | General recommendation method and system based on object tags |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107885774A (en) * | 2017-09-29 | 2018-04-06 | 北京京东尚科信息技术有限公司 | Data processing method and system |
| CN107885774B (en) * | 2017-09-29 | 2020-11-20 | 北京京东尚科信息技术有限公司 | Data processing method and system |
| CN109460519A (en) * | 2018-12-28 | 2019-03-12 | 上海晶赞融宣科技有限公司 | Browse object recommendation method and device, storage medium, server |
| CN111507768A (en) * | 2020-04-17 | 2020-08-07 | 腾讯科技（深圳）有限公司 | Determination method of potential user, model training method and related device |
| CN111507768B (en) * | 2020-04-17 | 2023-04-07 | 腾讯科技（深圳）有限公司 | Potential user determination method and related device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105608111B (en) | 2019-02-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104199896B (en) | The video similarity of feature based classification is determined and video recommendation method | |
| US9489688B2 (en) | Method and system for recommending search phrases | |
| CN109559208A (en) | A kind of information recommendation method, server and computer-readable medium | |
| US8296356B2 (en) | Rating based on relationship | |
| US9251170B2 (en) | Method and system for item selection based on characteristic matching | |
| CN104317900A (en) | Multiattribute collaborative filtering recommendation method oriented to social network | |
| US20130218865A1 (en) | Systems and methods for identifying and analyzing internet users | |
| JP2013525921A (en) | Method, system, and apparatus for query based on vertical search | |
| CN103886001A (en) | Personalized commodity recommendation system | |
| CN104599160A (en) | Commodity recommendation method and commodity recommendation device | |
| CN104268664A (en) | Method and device for recommending carpooling route | |
| CN103745100A (en) | Item-based explicit and implicit feedback mixing collaborative filtering recommendation algorithm | |
| CN105608111A (en) | Method and system for recommending object to terminal user | |
| WO2007101278A2 (en) | Behavioral trust rating filtering system | |
| CN110335112A (en) | A kind of Products Show method, apparatus and computer storage medium | |
| CN110111167A (en) | A kind of method and apparatus of determining recommended | |
| JP2013517561A (en) | Attribute aggregation for standard product units | |
| CN105894310A (en) | Personalized recommendation method | |
| Dev et al. | Recommendation system for big data applications based on set similarity of user preferences | |
| CN106027507A (en) | Method for recognizing anonymous user identity in social network | |
| EP2316063A1 (en) | Method and apparatus for providing data statistics | |
| CN109615466A (en) | The mixed method of commending contents and collaborative filtering recommending towards mobile ordering system | |
| CN106951459A (en) | Improvement Collaborative Filtering Recommendation Algorithm based on Information Entropy | |
| CN106599074A (en) | Method for calculating user time correlation similarity in collaborative filtering recommendation system | |
| CN106547919B (en) | A kind of distributed recommendation method of massive digital information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |