CN106779985A - A kind of method and system of personalized commercial sequence - Google Patents
A kind of method and system of personalized commercial sequence Download PDFInfo
- Publication number
- CN106779985A CN106779985A CN201710101299.3A CN201710101299A CN106779985A CN 106779985 A CN106779985 A CN 106779985A CN 201710101299 A CN201710101299 A CN 201710101299A CN 106779985 A CN106779985 A CN 106779985A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- collection
- cluster
- dimension
- 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.)
- Pending
Links
Classifications
-
- 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
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- Finance (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of method and system of personalized commercial sequence, it is characterised in that it comprises the following steps:S1:The information of collection crowd dimension and commodity dimension, cluster analysis is carried out to user's dimension and commodity dimension, obtains the cluster commodity collection containing crowd's information;S2:It is current crowd's partition clustering commodity collection;S3:The behavioral characteristic of user is obtained, the characteristic set of user is built, the affiliated crowd of user is determined;S4:It is user's selection cluster commodity collection, using dualistic logistic regression mode, separates the cluster commodity collection of user preference and non-preference, the cluster commodity collection of user preference is forward;S5:Sorted by score after pressing the commodity in the cluster commodity collection of user preference the exposure clicking rate of commodity, click on conversion ratio and commodity price COMPREHENSIVE CALCULATING, commodity are shown in order.The present invention selects commodity collection for user, and commodity collection internal product is sorted, and as far as possible can preferentially show the working days of user preference, more accurately draws a circle to approve the commodity collection of user preference.
Description
Technical field
The present invention relates to big data analysis field, more particularly to a kind of method and system of personalized commercial sequence.
Background technology
The Commodity Flow exhibition method of existing special selling App, the main working days feature according to commodity, daily special time issue is new
Product, attract user's upper new commodity of concern daily, and the support without search, is mainly main product feature with new product panic buying.With
Special selling commodity are enriched constantly, and the up to thousands of moneys of upper new commodity of daily Commodity Flow displaying simultaneously cover clothes, footwear bag, house, U.S.
The multiple classification such as food, mother and baby, but the commodity that user usually can patiently see only account for 10% less than substantial amounts of commodity are due to position
After resting against, inconvenience is checked, cause light exposure very low, the real commodity interested of many users are difficult to be found, and limit entirety
The growth of sales volume.
For problem above, the settling mode that special selling App ends generally use has:
(1) user's sex, age bracket (it is after 90s/85 after/it is after 80s/70 after etc.), user region, the master of the user such as consuming capacity
Want feature to distinguish crowd, different Commodity Flow lists are shown to different crowd.
(2) operation personnel according to the feature of commodity working days, hand picking goes out the commodity that may bring high conversion, by these
Commodity sequence shifts to an earlier date, and according to the overall sales volume situation of classification, advantage classification is preposition, and classification is combined intersection exhibition
Show.
(3) in fixed hole position embedded advertisement activity bit, by movable preferential lifting working days sales volume.
Existing solution major defect is:
(1) crowd's limited amount that principal character according to user itself is distinguished, can be typically divided between man, Ms,
It is after 80s, after 70, the main population such as high consumption, but because the commodity working days are up to thousands of, in face of daily mass users up to a million,
Only only small amounts of several crowds are not well positioned to meet the preference of user.
(2) it is artificial to carry out the working days and select and commodity sequence, the on the one hand daily fortune of this mode are intervened according to classification feature
Battalion's workload is larger, and on the other hand, due to runing the energy power restriction of itself, the working days quality selected is also uneven, causes effect
Fruit is often barely satisfactory.
(3) advertising campaign fixed bit is limited, and hoisting power is not high, carrys out effect to Commodity Flow entirety sales volume elevator belt and fails to understand
It is aobvious.
The content of the invention
The invention aims to overcome the shortcomings of above-mentioned background technology, there is provided a kind of side of personalized commodity sequence
Method and system, can better meet user preference, and the commodity of the preference of user are preferentially shown.
A kind of method of personalized commercial sequence that the present invention is provided, it comprises the following steps:
S1:The information of collection crowd dimension and commodity dimension, cluster analysis is carried out to user's dimension and commodity dimension, is obtained
Cluster commodity collection containing crowd's information;
S2:It is current crowd's partition clustering commodity collection;
S3:The behavioral characteristic of user is obtained, the characteristic set of user is built, the affiliated crowd of user is determined;
S4:It is user's selection cluster commodity collection, using dualistic logistic regression mode, separates gathering for user preference and non-preference
Class commodity collection, the cluster commodity collection of user preference is forward;
S5:Commodity in the cluster commodity collection of user preference are pressed into the exposure clicking rate of commodity, conversion ratio and commodity are clicked on
Sorted by score after price COMPREHENSIVE CALCULATING, commodity are shown in order.
A kind of system of personalized commercial sequence, it is characterised in that:It includes:
User's merchandise news is collected and Cluster Analysis module:Information for collecting crowd's dimension and commodity dimension, to
Family dimension and commodity dimension carry out cluster analysis, obtain the cluster commodity collection containing crowd's information;
Segregation class commodity collection division module:It is current crowd's partition clustering commodity collection;
User behavior feature collection module:Behavioral characteristic for obtaining user, builds the characteristic set of user, it is determined that with
The affiliated crowd in family;
Cluster commodity set analysis module:For being user's selection cluster commodity collection, using dualistic logistic regression mode, separate
The cluster commodity collection of user preference and non-preference, the cluster commodity collection of user preference is forward;
Cluster commodity collection internal sort module:The exposure that commodity in the cluster commodity collection of user preference press commodity is clicked on
Sorted by score after rate, click conversion ratio and commodity price COMPREHENSIVE CALCULATING, commodity are shown in order.
Beneficial effect:The method and system of a kind of personalized commodity sequence that the present invention is provided collect user's dimension and business
The information of product dimension, is user's selection commodity collection, and commodity collection internal product is sorted, can as far as possible by user preference
Working days preferentially show that more accurately the commodity collection of delineation user preference, lifts Consumer's Experience, so as to bring the pin of overall Commodity Flow
Amount goes up.
Brief description of the drawings
The method and system of a kind of personalized commodity sequence that Fig. 1 is provided for the present invention provided in an embodiment of the present invention
Method flow diagram.
Specific embodiment
Below in conjunction with the accompanying drawings and embodiment the present invention will be described in detail, the description of this part is only exemplary and explains
Property, there should not be any restriction effect to protection scope of the present invention.
As shown in figure 1, embodiment of the present invention provides a kind of method of personalized commercial sequence, it comprises the following steps:
S1:The information of collection crowd dimension and commodity dimension, cluster analysis is carried out to user's dimension and commodity dimension, is obtained
Cluster commodity collection containing crowd's information;
S2:It is current crowd's partition clustering commodity collection;
S3:The behavioral characteristic of user is obtained, the characteristic set of user is built, the affiliated crowd of user is determined;
S4:It is user's selection cluster commodity collection, using dualistic logistic regression mode, separates gathering for user preference and non-preference
Class commodity collection, the cluster commodity collection of user preference is forward;
S5:Commodity in the cluster commodity collection of user preference are pressed into the exposure clicking rate of commodity, conversion ratio and commodity are clicked on
Sorted by score after price COMPREHENSIVE CALCULATING, commodity are shown in order.
Specifically, the information of user's dimension includes:User's sex, age bracket, user region, consuming capacity;Institute's art
Commodity dimension includes classification, brand, price range, style, style.
Specifically, the method for the cluster is:The crowd's dimensional information and commodity dimensional information being first collected into are converted into
First commodity collection information, the first commodity collection information is converted into SimHash values, according to SimHash value difference values be used as cluster away from
From metric parameter, cluster commodity collection is drawn using k-means clustering algorithms, final number of clusters can be according to clustering convergence degree
It is specific quantitative.Specifically, such as the crowd characteristic of user is:After Ms+70+surplus one-piece dress+mother-infant breast-feeding phase+Beijing+disappear
Take ability 300~500, the first commodity collection information is as follows, and its form is<Feature 1, feature 2, two weights of feature association>,<
Ms, after 70,5>、<Ms, surplus one-piece dress, 4>、<Ms, mother-infant breast-feeding phase, 5>、<Ms, Beijing, 6>、<Ms, 300-
500,6>、<After 70, surplus one-piece dress, 3>、<After 70, mother-infant breast-feeding phase, 3>、<After 70, Beijing, 3>、<After 70,300-500,3
>、<Surplus one-piece dress, mother-infant breast-feeding phase, 2>、<Surplus one-piece dress, Beijing, 2>、<Surplus one-piece dress, 300-500,2>、<Mother and baby
Nursing period, Beijing, 1>、<Mother-infant breast-feeding phase, 300-500,1>、<Beijing, 300-500,1>, because length is limited, here using letter
The form of change, during above-mentioned first commodity collection information projected into the binary coding of 8 bit lengths, specific computing mode is as follows:
(1) a kind of computing mode of hash values, such as the SimHash methods in Java in HashMap, according to this side are specified
Method produces 8 binary codes, such as<Ms, after 70>SimHash values be 10001100,<Ms, surplus one-piece dress>
SimHash values be 11001001.
(2) above-mentioned 8 binary codings are multiplied by corresponding weight, (wherein 0 in multiplying to be converted into new value
For -1), such as<Ms, after 70>It is 5, -5, -5, -5,5,5, -5, -5 by ranking operation,<Ms, surplus one-piece dress>It is logical
It is 4,4, -4, -4,4, -4, -4,4 to cross ranking operation.
(3) value after all characteristic weighings is added and is merged, such as<Ms, after 70>With<Ms, surplus one-piece dress>Plus
Power is 9, -1, -9, -9,9,1, -9, -1 after merging, and then does dimension-reduction treatment, switchs to 1 more than 0, switching to less than or equal to 0
0, when 8 last SimHash results 1,0,0,0,1,1,0,0.
(4) two Simhash correspondences binary system (01 string) different quantity of value are the difference of the two Simhash values.
It is exemplified below:10101 have first, the 4th, the 5th different, the then difference of Simhash values successively with 00110 since first
Be worth is 3.
Specifically, it is that the method that current crowd's division commodity integrate is priority match classification scope, then in each class now
The degree of correlation commodity high such as selection age characteristics, price grade, region.
Specifically, the exposure clicking rate by commodity, click on conversion ratio and commodity price comprehensive calculation method is:Extract
CTR, CVR of each commodity, and smoothing processing is done, the smoothing processing formula is r=(C+alpha)/(I+alpha+beta)
(wherein C is that CTR or CVR, I are displaying number of times, and alpha, beta are smoothing factor), then according to formula:W=CTR*CVR*
Each commodity is given a mark by commodity price, and commodity are ranked up according to scoring event, and the purpose of formula is turned to be lifted
The weight of rate, exposure clicking rate commodity score higher, and prominent high price commodity, lifts entirety GMV outputs.
The embodiment of the present invention additionally provides a kind of system of personalized commercial sequence, and it includes:
User's merchandise news is collected and Cluster Analysis module:Information for collecting crowd's dimension and commodity dimension, to
Family dimension and commodity dimension carry out cluster analysis, obtain the cluster commodity collection containing crowd's information;
Cluster commodity collection division module:It is current crowd's partition clustering commodity collection;
User behavior feature collection module:Behavioral characteristic for obtaining user, builds the characteristic set of user, it is determined that with
The affiliated crowd in family;
Cluster commodity set analysis module:For being user's selection cluster commodity collection, using dualistic logistic regression mode, separate
The cluster commodity collection of user preference and non-preference, the cluster commodity collection of user preference is forward;
Cluster commodity collection internal sort module:The exposure that commodity in the cluster commodity collection of user preference press commodity is clicked on
Sorted by score after rate, click conversion ratio and commodity price COMPREHENSIVE CALCULATING, commodity are shown in order.
Specifically, user's merchandise news is collected and Cluster Analysis module includes:
The information module of user's dimension, for collecting user's sex, age bracket, user region, consuming capacity;
Commodity dimensional information collection module:For collecting commodity classification, brand, price range, style, style.
Specifically, user's merchandise news is collected and Cluster Analysis module includes Cluster Analysis module, the cluster point
Analysis module is used to for user's dimension and commodity dimension to carry out cluster analysis, and the method for cluster analysis is:The crowd being first collected into
Dimensional information and commodity dimensional information are converted into the first commodity collection information, and the first commodity collection information is converted into SimHash values, according to
SimHash value difference values are used as the distance metric parameter of cluster, draw cluster commodity collection using k-means clustering algorithms, finally
Number of clusters can be specifically quantitative according to clustering convergence degree, and specifically, such as the crowd characteristic of user is:After Ms+70+long
Money one-piece dress+mother-infant breast-feeding phase+Beijing+consuming capacity 300~500, the first commodity collection information is as follows, and its form is<Feature 1,
Feature 2, two weights of feature association>,<Ms, after 70,5>、<Ms, surplus one-piece dress, 4>、<Ms, the mother-infant breast-feeding phase,
5>、<Ms, Beijing, 6>、<Ms, 300-500,6>、<After 70, surplus one-piece dress, 3>、<After 70, mother-infant breast-feeding phase, 3>、<70
Afterwards, Beijing, 3>、<After 70,300-500,3>、<Surplus one-piece dress, mother-infant breast-feeding phase, 2>、<Surplus one-piece dress, Beijing, 2>、<It is long
Money one-piece dress, 300-500,2>、<Mother-infant breast-feeding phase, Beijing, 1>、<Mother-infant breast-feeding phase, 300-500,1>、<Beijing, 300-500,
1>, because length is limited, here in the form of simplification, the binary system that above-mentioned first commodity collection information projects to 8 bit lengths is compiled
In code, specific computing mode is as follows:
(1) a kind of computing mode of hash values, such as the SimHash methods in Java in HashMap, according to this side are specified
Method produces 8 binary codes, such as<Ms, after 70>SimHash values be 10001100,<Ms, surplus one-piece dress>
SimHash values be 11001001.
(2) above-mentioned 8 binary codings are multiplied by corresponding weight, (wherein 0 in multiplying to be converted into new value
For -1), such as<Ms, after 70>It is 5, -5, -5, -5,5,5, -5, -5 by ranking operation,<Ms, surplus one-piece dress>It is logical
It is 4,4, -4, -4,4, -4, -4,4 to cross ranking operation.
(3) value after all characteristic weighings is added and is merged, such as<Ms, after 70>With<Ms, surplus one-piece dress>Plus
Power is 9, -1, -9, -9,9,1, -9, -1 after merging, and then does dimension-reduction treatment, switchs to 1 more than 0, switching to less than or equal to 0
0, when 8 last SimHash results 1,0,0,0,1,1,0,0.
(4) two Simhash correspondences binary system (01 string) different quantity of value are the difference of the two Simhash values.
It is exemplified below:10101 have first, the 4th, the 5th different, the then difference of Simhash values successively with 00110 since first
Be worth is 3.
Specifically, the method that the specifically chosen commodity of commodity collection selecting module integrate is priority match classification scope, then
The degrees of correlation such as age characteristics, price grade, region commodity high are selected now in each class.
Specifically, the commodity collection internal sort module presses the exposure clicking rate of product, clicks on conversion ratio and commodity price
Comprehensive calculation method is:CTR, CVR of each commodity are extracted, and does smoothing processing, the smoothing processing formula is r=(C+
Alpha)/(I+alpha+beta) (wherein C is that CTR or CVR, I are displaying number of times, and alpha, beta are smoothing factor), so
Afterwards according to formula:Each commodity is given a mark by W=CTR*CVR* commodity prices, and commodity are ranked up according to scoring event,
The purpose of formula is to lift the weight of conversion ratio, exposure clicking rate commodity score higher, and prominent high price commodity, carrying
Rise entirety GMV outputs.
The method and system of a kind of personalized commodity sequence that the present invention is provided collect user's dimension and commodity dimension
Information, is user's selection commodity collection, and commodity collection internal product is sorted, can be preferential by the working days of user preference as far as possible
Displaying, more accurately draws a circle to approve the commodity collection of user preference, and commodity price power is highlighted in lifting Consumer's Experience, and sortord
Weight goes up so as to the sales volume for bringing overall Commodity Flow.
Obviously, those skilled in the art can carry out various changes and modification without deviating from essence of the invention to the present invention
God and scope.So, if these modifications of the invention and modification belong to the scope of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to comprising these changes and modification.
Claims (10)
1. a kind of method that personalized commercial sorts, it is characterised in that:It comprises the following steps:
S1:The information of collection crowd dimension and commodity dimension, cluster analysis is carried out to user's dimension and commodity dimension, is contained
The cluster commodity collection of crowd's information;
S2:It is current crowd's partition clustering commodity collection;
S3:The behavioral characteristic of user is obtained, the characteristic set of user is built, the affiliated crowd of user is determined;
S4:It is user's selection cluster commodity collection, using dualistic logistic regression mode, separates the cluster business of user preference and non-preference
Product collection, the cluster commodity collection of user preference is forward;
S5:Commodity in the cluster commodity collection of user preference are pressed into the exposure clicking rate of commodity, conversion ratio and commodity price is clicked on
Sorted by score after COMPREHENSIVE CALCULATING, commodity are shown in order.
2. the method that personalized commercial as claimed in claim 1 sorts, it is characterised in that:The packet of user's dimension
Include:User's sex, age bracket, user region, consuming capacity;Institute art commodity dimension include classification, brand, price range, style,
Style.
3. the method that personalized commercial as claimed in claim 1 sorts, it is characterised in that:The method of the cluster is:First will
The crowd's dimensional information and commodity dimensional information being collected into are converted into the first commodity collection information, and the first commodity collection information is converted into
SimHash values, the distance metric parameter of cluster is used as according to SimHash value difference values, is drawn using k-means clustering algorithms poly-
Class commodity collection, final number of clusters can be specifically quantitative according to clustering convergence degree.
4. the method that personalized commercial as claimed in claim 1 sorts, it is characterised in that:For current crowd divides commodity collection
Method is priority match classification scope, then selects the degrees of correlation such as age characteristics, price grade, region high now in each class
Commodity.
5. the method that personalized commercial as claimed in claim 1 sorts, it is characterised in that:The exposure by commodity is clicked on
Rate, click conversion ratio and commodity price comprehensive calculation method are:CTR, CVR of each commodity are extracted, and does smoothing processing, it is described
Smoothing processing formula be r=(C+alpha)/(I+alpha+beta) (wherein C is that CTR or CVR, I are displaying number of times,
Alpha, beta are smoothing factor), then according to formula:Each commodity is given a mark by W=CTR*CVR* commodity prices, root
Commodity are ranked up according to scoring event.
6. the system that a kind of personalized commercial sorts, it is characterised in that:It includes:
User's merchandise news is collected and Cluster Analysis module:Information for collecting crowd's dimension and commodity dimension, ties up to user
Degree and commodity dimension carry out cluster analysis, obtain the cluster commodity collection containing crowd's information;
Cluster commodity collection division module:It is current crowd's partition clustering commodity collection;
User behavior feature collection module:Behavioral characteristic for obtaining user, builds the characteristic set of user, determines user institute
Category crowd;
Cluster commodity set analysis module:For being user's selection cluster commodity collection, using dualistic logistic regression mode, user is separated
The cluster commodity collection of preference and non-preference, the cluster commodity collection of user preference is forward;
Cluster commodity collection internal sort module:By the commodity in the cluster commodity collection of user preference press commodity exposure clicking rate,
Sorted by score after clicking on conversion ratio and commodity price COMPREHENSIVE CALCULATING, commodity are shown in order.
7. the system that personalized commercial as claimed in claim 1 sorts, it is characterised in that:User's merchandise news collect and
Cluster Analysis module includes:
The information module of user's dimension, for collecting user's sex, age bracket, user region, consuming capacity;
Commodity dimensional information collection module:For collecting commodity classification, brand, price range, style, style.
8. the system that personalized commercial as claimed in claim 1 sorts, it is characterised in that:User's merchandise news collect and
Cluster Analysis module includes Cluster Analysis module, and the Cluster Analysis module is used to be clustered user's dimension and commodity dimension
Analyze, the method for cluster analysis is:The crowd's dimensional information and commodity dimensional information being first collected into are converted into the first commodity collection
Information, the first commodity collection information is converted into SimHash values, and the distance metric parameter of cluster is used as according to SimHash value difference values,
Cluster commodity collection is drawn using k-means clustering algorithms, final number of clusters can be specifically quantitative according to clustering convergence degree.
9. the system that personalized commercial as claimed in claim 1 sorts, it is characterised in that:The commodity collection selecting module is specific
The method that selection commodity integrate is priority match classification scope, then selects age characteristics, price grade, region now in each class
The commodity high etc. the degree of correlation.
10. the system that personalized commercial as claimed in claim 1 sorts, it is characterised in that:The commodity collection internal sort mould
Block by product exposure clicking rate, click on conversion ratio and commodity price comprehensive calculation method and be:Extract each commodity CTR,
CVR, and do smoothing processing, the smoothing processing formula be r=(C+alpha)/(I+alpha+beta) (wherein C be CTR or
CVR, I are displaying number of times, and alpha, beta are smoothing factor), then according to formula:W=CTR*CVR* commodity prices, to each
Commodity are given a mark, and commodity are ranked up according to scoring event.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710101299.3A CN106779985A (en) | 2017-02-24 | 2017-02-24 | A kind of method and system of personalized commercial sequence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710101299.3A CN106779985A (en) | 2017-02-24 | 2017-02-24 | A kind of method and system of personalized commercial sequence |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106779985A true CN106779985A (en) | 2017-05-31 |
Family
ID=58960815
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710101299.3A Pending CN106779985A (en) | 2017-02-24 | 2017-02-24 | A kind of method and system of personalized commercial sequence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106779985A (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107895296A (en) * | 2017-11-01 | 2018-04-10 | 武汉炎阳在线科技有限公司 | A kind of automated auction method and system |
CN107895299A (en) * | 2017-11-08 | 2018-04-10 | 浙江执御信息技术有限公司 | The exposure sort method and device of a kind of commodity |
CN108304512A (en) * | 2018-01-19 | 2018-07-20 | 北京奇艺世纪科技有限公司 | A kind of thick sort method of video search engine, device and electronic equipment |
CN109274987A (en) * | 2018-08-30 | 2019-01-25 | 武汉斗鱼网络科技有限公司 | A kind of video collection sort method, server and readable storage medium storing program for executing |
CN109903111A (en) * | 2017-12-11 | 2019-06-18 | 北京京东尚科信息技术有限公司 | For the sort method of personalized recommendation, order models training method and ordering system |
CN110288433A (en) * | 2019-06-12 | 2019-09-27 | 达疆网络科技(上海)有限公司 | One kind being based on clicking rate model prediction O2O real time individual sort method |
CN110378770A (en) * | 2019-07-25 | 2019-10-25 | 上海尊溢商务信息咨询有限公司 | A kind of method and device that e-commerce operation selects money to recommend |
CN110769283A (en) * | 2019-10-31 | 2020-02-07 | 广州市网星信息技术有限公司 | Video pushing method and device, computer equipment and storage medium |
CN111210298A (en) * | 2019-12-24 | 2020-05-29 | 武汉诸相网络科技有限公司 | Method and system for intelligently sorting commodity shelf based on big data processing c2c platform |
CN111435514A (en) * | 2019-01-15 | 2020-07-21 | 北京京东尚科信息技术有限公司 | Feature calculation method and device, sorting method and device, and storage medium |
CN111768218A (en) * | 2019-04-15 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for processing user interaction information |
CN112381607A (en) * | 2020-11-12 | 2021-02-19 | 杭州时趣信息技术有限公司 | Network commodity ordering method, device, equipment and medium |
CN113450167A (en) * | 2020-03-25 | 2021-09-28 | 北京沃东天骏信息技术有限公司 | Commodity recommendation method and device |
CN113744015A (en) * | 2020-10-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Sorting method, device, equipment and computer storage medium |
CN118195690A (en) * | 2024-05-15 | 2024-06-14 | 浙江卡赢信息科技有限公司 | Intelligent commodity shelf selection method for point exchange mall based on bank user |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102254028A (en) * | 2011-07-22 | 2011-11-23 | 青岛理工大学 | Personalized commodity recommendation method and system integrating attributes and structural similarity |
CN106326483A (en) * | 2016-08-31 | 2017-01-11 | 华南理工大学 | Collaborative recommendation method with user context information aggregation |
-
2017
- 2017-02-24 CN CN201710101299.3A patent/CN106779985A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102254028A (en) * | 2011-07-22 | 2011-11-23 | 青岛理工大学 | Personalized commodity recommendation method and system integrating attributes and structural similarity |
CN106326483A (en) * | 2016-08-31 | 2017-01-11 | 华南理工大学 | Collaborative recommendation method with user context information aggregation |
Non-Patent Citations (3)
Title |
---|
袁利: "基于聚类的协同过滤个性化推荐算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑(月刊)》 * |
赵晓永: "《面向云计算的数据存储关键技术研究》", 31 December 2014 * |
陈树良: "《统计软件》", 30 November 2014 * |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107895296A (en) * | 2017-11-01 | 2018-04-10 | 武汉炎阳在线科技有限公司 | A kind of automated auction method and system |
CN107895299A (en) * | 2017-11-08 | 2018-04-10 | 浙江执御信息技术有限公司 | The exposure sort method and device of a kind of commodity |
CN107895299B (en) * | 2017-11-08 | 2020-12-01 | 浙江执御信息技术有限公司 | Commodity exposure sequencing method and device |
CN109903111A (en) * | 2017-12-11 | 2019-06-18 | 北京京东尚科信息技术有限公司 | For the sort method of personalized recommendation, order models training method and ordering system |
CN108304512A (en) * | 2018-01-19 | 2018-07-20 | 北京奇艺世纪科技有限公司 | A kind of thick sort method of video search engine, device and electronic equipment |
CN109274987A (en) * | 2018-08-30 | 2019-01-25 | 武汉斗鱼网络科技有限公司 | A kind of video collection sort method, server and readable storage medium storing program for executing |
CN111435514B (en) * | 2019-01-15 | 2024-04-09 | 北京京东尚科信息技术有限公司 | Feature calculation method and device, ranking method and device, and storage medium |
CN111435514A (en) * | 2019-01-15 | 2020-07-21 | 北京京东尚科信息技术有限公司 | Feature calculation method and device, sorting method and device, and storage medium |
CN111768218A (en) * | 2019-04-15 | 2020-10-13 | 北京沃东天骏信息技术有限公司 | Method and device for processing user interaction information |
CN110288433A (en) * | 2019-06-12 | 2019-09-27 | 达疆网络科技(上海)有限公司 | One kind being based on clicking rate model prediction O2O real time individual sort method |
CN110378770A (en) * | 2019-07-25 | 2019-10-25 | 上海尊溢商务信息咨询有限公司 | A kind of method and device that e-commerce operation selects money to recommend |
CN110769283A (en) * | 2019-10-31 | 2020-02-07 | 广州市网星信息技术有限公司 | Video pushing method and device, computer equipment and storage medium |
CN111210298A (en) * | 2019-12-24 | 2020-05-29 | 武汉诸相网络科技有限公司 | Method and system for intelligently sorting commodity shelf based on big data processing c2c platform |
CN113450167A (en) * | 2020-03-25 | 2021-09-28 | 北京沃东天骏信息技术有限公司 | Commodity recommendation method and device |
CN113744015A (en) * | 2020-10-20 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Sorting method, device, equipment and computer storage medium |
CN112381607B (en) * | 2020-11-12 | 2023-11-24 | 杭州时趣信息技术有限公司 | Network commodity ordering method, device, equipment and medium |
CN112381607A (en) * | 2020-11-12 | 2021-02-19 | 杭州时趣信息技术有限公司 | Network commodity ordering method, device, equipment and medium |
CN118195690A (en) * | 2024-05-15 | 2024-06-14 | 浙江卡赢信息科技有限公司 | Intelligent commodity shelf selection method for point exchange mall based on bank user |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106779985A (en) | A kind of method and system of personalized commercial sequence | |
CN107590675B (en) | User shopping behavior identification method based on big data, storage device and mobile terminal | |
AU2013222023A1 (en) | Marketing device, marketing method, program and recording medium | |
CN109255651A (en) | A kind of search advertisements conversion intelligent Forecasting based on big data | |
CN103345512A (en) | Online advertising click-through rate forecasting method and device based on user attribute | |
CN102591876A (en) | Sequencing method and device of search results | |
CN103309990A (en) | User multidimensional analysis and monitoring method based on public information of Internet user | |
CN101783004A (en) | Fast intelligent commodity recommendation system | |
CN106570718A (en) | Information releasing method and releasing system | |
JP6810745B2 (en) | Systems and methods for segmenting customers with mixed attribute types using target clustering techniques | |
CN107658001A (en) | A kind of home-use oily health control method and system | |
CN108876430A (en) | A kind of advertisement sending method based on crowd characteristic, electronic equipment and storage medium | |
CN107784510A (en) | Sales achievement statistical analysis system and method based on shops's retail terminal | |
CN111949711B (en) | Commodity recommendation system based on decision-making high-utility negative sequence rule mining and working method thereof | |
CN110378770A (en) | A kind of method and device that e-commerce operation selects money to recommend | |
CN112597398B (en) | Medicine recommendation model application method and system | |
CN109978061A (en) | A kind of recovery method as resource based on resource reclaim O2O platform | |
CN108053322A (en) | The customer investment return evaluation method and system of vehicle | |
CN109740624A (en) | A kind of logistics supply chain needing forecasting method based on big data | |
CN115496566B (en) | Regional specialty recommendation method and system based on big data | |
CN110084683A (en) | A kind of Intelligent Business processing system | |
CN106326318A (en) | Search method and device | |
CN106919995A (en) | A kind of method and device for judging user group's loss orientation | |
CN116308609A (en) | Commodity recommendation method and device | |
CN106600044A (en) | Method and apparatus for determining vehicle sales quantity prediction model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170531 |
|
RJ01 | Rejection of invention patent application after publication |