CN105321089A - Method and system for e-commerce recommendation based on multi-algorithm fusion - Google Patents
Method and system for e-commerce recommendation based on multi-algorithm fusion Download PDFInfo
- Publication number
- CN105321089A CN105321089A CN201410338542.XA CN201410338542A CN105321089A CN 105321089 A CN105321089 A CN 105321089A CN 201410338542 A CN201410338542 A CN 201410338542A CN 105321089 A CN105321089 A CN 105321089A
- Authority
- CN
- China
- Prior art keywords
- commodity
- user
- group
- behavior data
- similar
- 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
Abstract
The invention provides a method and a system for e-commerce recommendation based on multi-algorithm fusion, and belongs to the technical field of e-commerce. The method comprises the steps of S1, acquiring historical behavior data of users on an e-commerce website; S2, grouping the users according to purchasing preference of the historical behavior data of the users, and dividing users with similar purchasing preference into the same user group; S3, layering commodities according to commodity attributes, and corresponding commodities with the same commodity attribute to the same commodity layering catalogue; S4, constructing commodity recommendation rules according to grouping of the users and the commodity layering catalogues; and S5, recommending the commodities to the users according to the commodity recommendation rules. The technical scheme provided by the invention can improve the stability and the diversity of a commodity recommendation system through fusion of multiple recommendation algorithms.
Description
Technical field
The present invention relates to technical field of electronic commerce, particularly a kind of electronic commerce recommending method based on many algorithm fusions and system.
Background technology
Along with the universal of internet and the development of ecommerce; e-commerce system is while providing more and more selection for user; its structure also becomes more complicated; user often can get lost in a large amount of merchandise news spaces; the commodity that cannot oneself be found smoothly to need, thus the stability and the diversity that reduce commending system.
Summary of the invention
For the above-mentioned defect of prior art, technical matters to be solved by this invention is how by the fusion of multiple proposed algorithm, improves stability and the diversity of commercial product recommending system.
For achieving the above object, on the one hand, the invention provides a kind of electronic commerce recommending method based on many algorithm fusions, the method comprising the steps of:
The historical behavior data of user on step S1, collection e-commerce website;
Step S2, according to the purchase preference of the historical behavior data of user, user to be hived off, the user buying preference similar is divided in same customer group;
Step S3, according to the attribute of commodity, layering is carried out to commodity, under the commodity gradation directory corresponding identical by the commodity of same alike result;
Step S4, according to the hiving off of user, commodity gradation directory structure commercial product recommending rule;
Step S5, according to commercial product recommending rule, by commercial product recommending to user.
Preferably, described structure commercial product recommending rule comprises:
The correlativity of commodity is calculated according to historical behavior data;
Correlativity according to commodity retains and/or the similar commodity group of completion, structure commodity association table;
Associated articles is obtained according to described commodity association table.
Preferably, described structure commercial product recommending rule specifically comprises:
According to historical behavior data respectively to commodity three grades of catalogues and similar commodity group compute associations relation;
The Relevance scores between each similar commodity group is calculated according to commodity three grades of catalogues and similar commodity group;
The similar commodity group that similarity is greater than setting threshold value is retained in each similar commodity group;
When the associated articles group quantity of similar commodity group is less than the minimum association similar commodity group threshold value of setting, according to commodity three grades of catalogues of association, the commodity group of getting sales volume under associated articles three grades of catalogues the highest carries out completion to this similar commodity group;
According to the commodity association table of structure, determine similar commodity group belonging to each commodity, get the similar commodity group of this similar commodity group association, according to the associated articles of the highest commodity of sales volume as these commodity under each commodity association group.
Preferably, also comprise after gathering the historical behavior data of user on e-commerce website:
By the abnormal data in the historical behavior data of the user collected being filtered based on seasonal effect in time series algorithm.
Preferably, described abnormal data carries out filtration and comprises:
Buy the kind quantity of commodity and the commodity sum of each kind commodity purchasing by limited subscriber, filter out abnormal buying behavior data;
According to the time series that the history sales volume of commodity is noted down, detect commercial promotions abnormal data;
Described abnormal data is processed.
Preferably, described commercial promotions abnormal information to be detected, and the smoothing process of abnormal sales data is specifically comprised:
According to the time series that the history sales volume of commodity is noted down, detect commercial promotions abnormal data;
To the smoothing process of described abnormal data.
Preferably, described according to commercial product recommending rule, commercial product recommending is specifically comprised to user:
Hiving off according to user, obtain this hive off under commodity association table, the commodity corresponding according to user behavior, according to the associated score between commodity from contingency table, select commercial product recommending that associated score is high to user.
On the other hand, the present invention also provides a kind of Technologies of Recommendation System in E-Commerce based on many algorithm fusions, and this system comprises:
User behavior data acquisition module, for gathering the historical behavior data of user;
Tenant group module, the purchase preference for the historical behavior data according to user is hived off to user, is divided in same customer group by the user buying preference similar;
Commodity hierarchical block, carries out layering for the attribute according to commodity to commodity, under the commodity gradation directory corresponding identical by the commodity of same alike result;
Recommendation rules constructing module, for according to hiving off, commodity gradation directory structure commercial product recommending rule;
User's commercial product recommending module, for according to the commercial product recommending rule constructed, by commercial product recommending to user.
Preferably, this system also comprises:
User behavior data processing module, filters the abnormal data in the historical behavior data of the user collected based on seasonal effect in time series algorithm for passing through.
Preferably, described user behavior data processing module specifically comprises:
First behavior data processing module, the commodity for the kind quantity and each kind commodity purchasing of being bought commodity by limited subscriber are total, filter out abnormal buying behavior data;
Second behavioral data processing module, for detecting commercial promotions abnormal information, and to the smoothing process of abnormal sales data.
Preferably, described recommendation rules constructing module comprises:
Correlation calculations module, for calculating the correlativity of commodity according to historical behavior data;
Contingency table constructing module, retains and/or the similar commodity group of completion for the correlativity according to commodity, structure commodity association table;
Commodity acquisition module, for obtaining associated articles according to described commodity association table.
Preferably, described user's commercial product recommending module specifically for hiving off according to user, obtain this hive off under commodity association table, according to the associated score between commodity from commodity contingency table, select commercial product recommending that associated score is high to user.
Provided by the invention based in many algorithm fusions electronic commerce recommending method and system, the correlation rule of concept based layering is have employed respectively in this system, the collaborative filtering of concept based layering, filter, based on personalized recommendation and the Knowledge based engineering proposed algorithm of cluster based on seasonal effect in time series abnormity point.By the fusion of multiple proposed algorithm, while the stability ensureing commending system, improve the diversity of commending system, and the cold start-up problem of the commending system effectively solved.
Accompanying drawing explanation
Fig. 1 a is the schematic flow sheet of the electronic commerce recommending method based on many algorithm fusions in one embodiment of the present of invention;
Fig. 1 b is the structure process flow diagram of the historical data recommendation rules according to user in a preferred embodiment of the present invention;
Fig. 1 c is the schematic flow sheet of the user's Recommendations in a preferred embodiment of the present invention;
Fig. 1 d is the schematic flow sheet of the abnormal sales behavior detection and treatment method in a preferred embodiment of the present invention;
Fig. 2 is the structural representation of the Technologies of Recommendation System in E-Commerce based on many algorithm fusions in an alternative embodiment of the invention;
Fig. 3 is the commodity hierarchy schematic diagram in a preferred embodiment of the present invention.
Embodiment
For making those skilled in the art understand technical scheme of the present invention better, below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
In prior art; along with the universal of internet and the development of ecommerce, e-commerce system is while providing more and more selection for user, its structure also becomes more complicated; user often can get lost in a large amount of merchandise news spaces, the commodity that cannot oneself be found smoothly to need.In electronic commerce recommending method based on many algorithm fusions provided by the invention and system, by have employed the correlation rule of concept based layering, the collaborative filtering of concept based layering, filters, based on personalized recommendation and the Knowledge based engineering proposed algorithm of cluster based on seasonal effect in time series abnormity point.By the fusion of multiple proposed algorithm, improve stability and the diversity of commending system.
Fig. 1 is the schematic flow sheet of the electronic commerce recommending method based on many algorithm fusions in a preferred embodiment of the present invention, and as shown in Figure 1, the method comprises the steps:
The historical behavior data of user on step S101, collection e-commerce website;
Step S102, according to the purchase preference of the historical behavior data of user, user to be hived off, the user buying preference similar is divided in same customer group;
Step S103, according to the attribute of commodity, layering is carried out to commodity, under the commodity gradation directory corresponding identical by the commodity of same alike result;
Step S104, according to the hiving off of user, commodity gradation directory structure commercial product recommending rule;
Step S105, according to commercial product recommending rule, by commercial product recommending to user.
Preferably, described structure commercial product recommending rule comprises:
The correlativity of commodity is calculated according to historical behavior data;
Correlativity according to commodity retains and/or the similar commodity group of completion, structure commodity association table;
Associated articles is obtained according to described commodity association table.
Preferably, construct commercial product recommending rule specifically to comprise:
Step S1041, according to historical behavior data respectively to commodity three grades of catalogues and similar commodity group compute associations relation;
Step S1042, calculate the Relevance scores between each similar commodity group according to commodity three grades of catalogues and similar commodity group;
Step S1043, retain in each similar commodity group similarity be greater than setting threshold value similar commodity group;
Step S1044, when the associated articles group quantity of similar commodity group is less than the minimum association similar commodity group threshold value of setting, according to commodity three grades of catalogues of association, the commodity group of getting sales volume under associated articles three grades of catalogues the highest carries out completion to this similar commodity group;
Step S1045, commodity association table according to structure, determine similar commodity group belonging to each commodity, get the similar commodity group of this similar commodity group association, according to the associated articles of the highest commodity of sales volume as these commodity under each commodity association group.
Preferably, also comprise after gathering the historical behavior data of user on e-commerce website: by the abnormal data in the historical behavior data of the user collected being filtered based on seasonal effect in time series algorithm.
Preferably, abnormal data carries out filtration and comprises:
The commodity sum of step S1061, the kind quantity being bought commodity by limited subscriber and each kind commodity purchasing, filters out abnormal buying behavior data;
Step S1062, the time series noted down according to the history sales volume of commodity, detect commercial promotions abnormal data, and process described abnormal data.
Preferably, commercial promotions abnormal information is detected, and the smoothing process of abnormal sales data is specifically comprised: the time series of the history sales volume record of structure commodity; Described time series is divided into secular trend, periodically trend and residual error; To the smoothing process of the abnormal data that residual error is corresponding, eliminate residual error, thus eliminate the impact of abnormal sales data on structure commercial product recommending rule.
Preferably, according to commercial product recommending rule, specifically comprised by commercial product recommending: hiving off according to user to user, obtain this lower commodity association table that hives off, according to the associated score between commodity from commodity contingency table, the high commercial product recommending of selection associated score is to user.Concrete steps are as follows:
The log-on message of step S1051, acquisition user;
Step S1052, obtain tenant group information according to the log-on message of user;
The behavioural information of step S1053, acquisition user;
Step S1054, hiving off according to user, obtain this hive off under commodity association table, the commodity corresponding according to user behavior, according to the associated score between commodity from contingency table, select commercial product recommending that associated score is high to user.
Relevant technical staff in the field will be understood that, corresponding with method of the present invention, the present invention also comprises a kind of Technologies of Recommendation System in E-Commerce based on many algorithm fusions simultaneously, with said method step correspondingly, as shown in Figure 2, this system comprises: user behavior data acquisition module 201, tenant group module 202, commodity hierarchical block 203, recommendation rules constructing module 204 and user's commercial product recommending module 206.Wherein, user behavior data acquisition module 201, for gathering the historical behavior data of user, carries out the excavation of recommendation rules by user behavior data; Tenant group module 202 is hived off to user for the purchase preference of the historical behavior data according to user, is divided in same customer group by the user buying preference similar; Commodity hierarchical block 203 carries out layering for the attribute according to commodity to commodity, under the commodity gradation directory corresponding identical by the commodity of same alike result; Recommendation rules constructing module 204 is for according to the hiving off of user, commodity gradation directory structure commercial product recommending rule; User's commercial product recommending module 206 for according to commercial product recommending rule, by commercial product recommending to user.
Particularly, user behavior data acquisition module 201, for gathering the historical behavior data of user, comprises the webpage that user browses, browse the duration of webpage, the commodity of user search, user put into collection folder commodity, user adds the commodity of shopping cart, and user submits the commodity of order to.Preferably, user behavior data acquisition module 201 can also carry out recommendation rules excavation by obtaining user behavior data.
Preferably, this system also comprises: user behavior data processing module 205, based on seasonal effect in time series algorithm, the abnormal data in the historical behavior data of the user collected is filtered for passing through, thus filter out the abnormal buying behavior of user and the sales behavior of abnormal commodity.User behavior data processing module 205 is mainly used in realizing data cleansing, filtering data noise, thus improves the accuracy rate of recommending.User behavior data processing module 205 is cleaned data and is mainly comprised following two parts:
Part I: the abnormal buying behavior of user;
Part II: the abnormal merchandise related information brought due to sales promotion etc.
Preferably, user behavior data processing module 205 specifically comprises: the first behavior data processing module 2051 and the second behavioral data processing module 2052.Wherein, the first behavior data processing module 2051 is total for the commodity of the kind quantity and each kind commodity purchasing of being bought commodity by limited subscriber, filters out abnormal buying behavior data, realizes the abnormal buying behavior of process user; Second behavioral data processing module 2052 is for detecting commercial promotions abnormal information, and to the smoothing process of abnormal sales data, eliminate the impact of abnormal sales data on structure commercial product recommending rule, thus the abnormal merchandise related information that realization process brings due to sales promotion.
In the specific works process of user behavior data processing module 205, the idiographic flow of abnormal sales behavior detection and treatment is as follows:
The first step: obtain the record of commodity history sales volume;
Second step: to history sales volume structure time series;
3rd step: decompose time series, obtains the trend steady in a long-term of sales volume respectively, periodically trend and residual error, and what wherein residual error was corresponding is abnormal data;
4th step: the abnormal data corresponding to residual error is smoothing, thus eliminate residual error impact.
Particularly, tenant group module 202 for predicting the purchase preference of user, and is hived off to user according to the purchase preference of user, makes user in each colony buy preference and has similarity, and realize the differentiation type buying preference between colony.In actual applications, can according to the log-on message of user and behavioral data to the sex of user, at the age, the Long-term Interest of user, the short-term purchase intention of user is predicted, to hive off to user according to predicting the outcome.
Particularly, commodity hierarchical block 203, for carrying out layering according to merchandise classification to commodity, under making the gradation directory that similar commodity are corresponding identical, thus realizes the excavation of concept based layering recommendation rules.
Preferably, recommendation rules constructing module 204 excavates recommendation rules for realizing according to the historical data of user, is the nucleus module in Technologies of Recommendation System in E-Commerce.Recommendation rules constructing module 204 specifically comprises: correlation calculations module 2041, contingency table constructing module 2042 and commodity acquisition module 2043.Wherein, correlation calculations module 2041 is for calculating the correlativity of commodity according to historical behavior data; Contingency table constructing module 2042 retains and/or the similar commodity group of completion for the correlativity according to commodity, structure commodity association table; Commodity acquisition module 2043 is for obtaining associated articles according to described commodity association table.
Preferably, correlation calculations module 2041 comprises further: incidence relation computing module 2041 with associate points calculating module 2042.Wherein, incidence relation computing module 2041 for according to historical behavior data respectively to commodity three grades of catalogues and similar commodity group compute associations relation; Association points calculating module 2042 is for calculating the Relevance scores between each similar commodity group according to commodity three grades of catalogues and similar commodity group; Contingency table constructing module 2042 comprises further: similar commodity group retains module 2043 and similar commodity group completion module 2044.Wherein, similar commodity group retains module 2043 is greater than setting threshold value similar commodity group for retaining similarity in each similar commodity group; Similar commodity group completion module 2044 is for being less than the minimum association similar commodity group threshold value of setting during in the associated articles group quantity of similar commodity group, according to commodity three grades of catalogues of association, the commodity group of getting sales volume under associated articles three grades of catalogues the highest carries out completion to this similar commodity group; Commodity acquisition module 2043 comprises further: associated articles group acquisition module 2045, for the similar commodity group association list according to structure, determine similar commodity group belonging to each commodity, get the similar commodity group of this similar commodity group association, according to the associated articles of the highest commodity of sales volume as these commodity under each commodity association group.
Particularly, the specific works flow process of recommendation rules constructing module 204 is as follows:
The first step: first according to tenant group information, gets the historical data of each user inside the group;
Second step: according to the historical data of the user in each customer group respectively to commodity three grades of catalogues, similar commodity group compute associations relation, incidence relation calculates can adopt existing incidence relation algorithm, comprises correlation rule and collaborative filtering.
3rd step: between each similar commodity group associate be divided into the weighted accumulation that associates score and the three grades of catalogue scores of the commodity belonging to it between this similar commodity group and.Computing formula is as follows: Score=W1*SP_Score+W2*Parent_Score, and wherein, W1 is similar commodity group relevance weight, and W2 is the relevance weight of commodity three grades of catalogues, W1+W2=1; SP_Score is the correlativity between similar commodity group, and Parent_Score is the correlativity between commodity three grades of catalogues.Certainly, the present invention also can adopt other mark amalgamation modes, such as, similar commodity group score and commodity three grades of catalogue score products can be adopted to associate score as final score as each similar commodity group.
In the present invention, commodity three grades of catalogues and similar commodity group two levels are adopted to carry out calculating the correlativity between type of merchandize.Fig. 3 is the commodity hierarchy schematic diagram in a preferred embodiment of the present invention, as shown in Figure 3, commodity three grades of catalogues that the present invention adopts comprise: commodity root directory, the multiple three grades of catalogues under the multiple first class catalogues under commodity root directory, the multiple second-level directories under first class catalogue, second-level directory and the multiple similar commodity under three grades of catalogues.The present invention includes but be not limited to above two levels and calculate Relevance scores, such as, other level relevance scores cumulative also can be adopted as the Relevance scores between similar commodity.
4th step: the final similar commodity group obtaining similarity and be greater than threshold value is retained to each similar commodity group.
5th step: if the associated articles group quantity of similar commodity group is less than the minimum association similar commodity group threshold value of setting, then according to three grades of goods catalogue of association, the commodity group of getting sales volume under associated articles catalogue the highest carries out completion to this similar commodity group.
6th step: according to the similar commodity group contingency table of structure, first similar commodity group belonging to it is determined to each commodity, then get the similar commodity group of association of this similar commodity group, under each associated articles group according to the associated articles of the highest commodity of sales volume list write off amount as these commodity.
7th step: above-mentioned steps is adopted to each customer group, thus obtain the associated articles of each commodity under each customer group.
User's commercial product recommending module 206, for the recommendation rules constructed according to recommendation rules constructing module 204, realizes the commercial product recommending of user.Preferably, user's commercial product recommending module 206 specifically for hiving off according to user, obtain this hive off under commodity association table, according to the associated score between commodity from commodity contingency table, select commercial product recommending that associated score is high to user.Particularly, the specific works flow process of user's commercial product recommending module 206 is as follows:
The first step: obtain user login information;
Second step: obtain tenant group corresponding to user according to the Clustering information of user;
3rd step: the behavioural information obtaining user, as browsed, buying behavior etc.;
4th step: hive off belonging to user, obtain this hive off under commodity association table, browse according to user or the behavior commodity such as purchase, according to the associated score between commodity from contingency table, select commercial product recommending that associated score is high to user.
The present embodiment provide based in many algorithm fusions Technologies of Recommendation System in E-Commerce, have employed the correlation rule of concept based layering within the system respectively, the collaborative filtering of concept based layering, filter, based on personalized recommendation and the Knowledge based engineering proposed algorithm of cluster based on seasonal effect in time series abnormity point.By the fusion of multiple proposed algorithm, while the stability ensureing commending system, improve the diversity of commending system, and the cold start-up problem of the commending system effectively solved.
Be understandable that, the illustrative embodiments that above embodiment is only used to principle of the present invention is described and adopts, but the present invention is not limited thereto.For those skilled in the art, without departing from the spirit and substance in the present invention, can make various modification and improvement, these modification and improvement are also considered as protection scope of the present invention.
Claims (11)
1. based on an electronic commerce recommending method for many algorithm fusions, it is characterized in that, described method comprises step:
The historical behavior data of user on step S1, collection e-commerce website;
Step S2, according to the purchase preference of the historical behavior data of user, user to be hived off, the user buying preference similar is divided in same customer group;
Step S3, according to the attribute of commodity, layering is carried out to commodity, under the commodity gradation directory corresponding identical by the commodity of same alike result;
Step S4, according to the hiving off of user, commodity gradation directory structure commercial product recommending rule;
Step S5, according to commercial product recommending rule, by commercial product recommending to user.
2. method according to claim 1, is characterized in that, described structure commercial product recommending rule comprises:
The correlativity of commodity is calculated according to historical behavior data;
Correlativity according to commodity retains and/or the similar commodity group of completion, structure commodity association table;
Associated articles is obtained according to described commodity association table.
3. method according to claim 2, is characterized in that, described structure commercial product recommending rule specifically comprises:
According to historical behavior data respectively to commodity three grades of catalogues and similar commodity group compute associations relation;
The Relevance scores between each similar commodity group is calculated according to commodity three grades of catalogues and similar commodity group;
The similar commodity group that similarity is greater than setting threshold value is retained in each similar commodity group;
When the associated articles group quantity of similar commodity group is less than the minimum association similar commodity group threshold value of setting, according to commodity three grades of catalogues of association, the commodity group of getting sales volume under associated articles three grades of catalogues the highest carries out completion to this similar commodity group;
According to the commodity association table of structure, determine similar commodity group belonging to each commodity, get the similar commodity group of this similar commodity group association, according to the associated articles of the highest commodity of sales volume as these commodity under each commodity association group.
4. method according to claim 3, is characterized in that, also comprises after gathering the historical behavior data of user on e-commerce website:
By the abnormal data in the historical behavior data of the user collected being filtered based on seasonal effect in time series algorithm.
5. method according to claim 4, is characterized in that, described abnormal data carries out filtration and comprises:
Buy the kind quantity of commodity and the commodity sum of each kind commodity purchasing by limited subscriber, filter out abnormal buying behavior data;
According to the time series that the history sales volume of commodity is noted down, detect commercial promotions abnormal data, and described abnormal data is processed.
6. method according to claim 3, is characterized in that, described according to commercial product recommending rule, is specifically comprised by commercial product recommending to user:
Hiving off according to user, obtain this hive off under commodity association table, according to the associated score between commodity from commodity contingency table, select commercial product recommending that associated score is high to user.
7. based on a Technologies of Recommendation System in E-Commerce for many algorithm fusions, it is characterized in that, comprising:
User behavior data acquisition module, for gathering the historical behavior data of user;
Tenant group module, the purchase preference for the historical behavior data according to user is hived off to user, is divided in same customer group by the user buying preference similar;
Commodity hierarchical block, carries out layering for the attribute according to commodity to commodity, under the commodity gradation directory corresponding identical by the commodity of same alike result;
Recommendation rules constructing module, for according to the hiving off of user, commodity gradation directory structure commercial product recommending rule;
User's commercial product recommending module, for according to commercial product recommending rule, by commercial product recommending to user.
8. system according to claim 7, is characterized in that, also comprises:
User behavior data processing module, filters the abnormal data in the historical behavior data of the user collected based on seasonal effect in time series algorithm for passing through.
9. system according to claim 8, is characterized in that, described user behavior data processing module specifically comprises:
First behavior data processing module, the commodity for the kind quantity and each kind commodity purchasing of being bought commodity by limited subscriber are total, filter out abnormal buying behavior data;
Second behavioral data processing module, for the time series noted down according to the history sales volume of commodity, detects commercial promotions abnormal data, processes described abnormal data.
10. system according to claim 8, is characterized in that, described recommendation rules constructing module comprises:
Correlation calculations module, for calculating the correlativity of commodity according to historical behavior data;
Contingency table constructing module, retains and/or the similar commodity group of completion for the correlativity according to commodity, structure commodity association table;
Commodity acquisition module, for obtaining associated articles according to described commodity association table.
11. systems according to claim 7, it is characterized in that, described user's commercial product recommending module specifically for hiving off according to user, obtain this hive off under commodity association table, according to the associated score between commodity from commodity contingency table, select commercial product recommending that associated score is high to user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410338542.XA CN105321089A (en) | 2014-07-16 | 2014-07-16 | Method and system for e-commerce recommendation based on multi-algorithm fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410338542.XA CN105321089A (en) | 2014-07-16 | 2014-07-16 | Method and system for e-commerce recommendation based on multi-algorithm fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105321089A true CN105321089A (en) | 2016-02-10 |
Family
ID=55248425
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410338542.XA Pending CN105321089A (en) | 2014-07-16 | 2014-07-16 | Method and system for e-commerce recommendation based on multi-algorithm fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105321089A (en) |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105761111A (en) * | 2016-02-22 | 2016-07-13 | 青岛海尔股份有限公司 | Information processing method and information processing device based on dosage of ingredients |
| CN106910108A (en) * | 2017-01-24 | 2017-06-30 | 武汉奇米网络科技有限公司 | A kind of items list methods of exhibiting and system |
| CN107093120A (en) * | 2016-09-28 | 2017-08-25 | 北京小度信息科技有限公司 | Service strategy method for pushing and device |
| CN107358508A (en) * | 2017-07-18 | 2017-11-17 | 广州优视网络科技有限公司 | A kind of similar article item management method and device |
| CN107402742A (en) * | 2017-08-04 | 2017-11-28 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN107657506A (en) * | 2017-09-06 | 2018-02-02 | 北京五八到家信息技术有限公司 | The intelligent recommendation method and commending system of a kind of service commodity |
| CN107679905A (en) * | 2017-09-28 | 2018-02-09 | 千寻位置网络有限公司 | Market guidance method and system, service terminal, memory based on customer location |
| CN107730336A (en) * | 2016-08-12 | 2018-02-23 | 苏宁云商集团股份有限公司 | Commodity method for pushing and device in a kind of online transaction |
| CN107730337A (en) * | 2016-08-12 | 2018-02-23 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN107918882A (en) * | 2016-10-11 | 2018-04-17 | 阿里巴巴集团控股有限公司 | The method and device of data object information is provided |
| CN108009867A (en) * | 2016-10-28 | 2018-05-08 | 百度在线网络技术（北京）有限公司 | Information output method and device |
| CN108205768A (en) * | 2016-12-20 | 2018-06-26 | 百度在线网络技术（北京）有限公司 | Database building method and data recommendation method and device, equipment and storage medium |
| CN108446963A (en) * | 2018-03-30 | 2018-08-24 | 深圳春沐源控股有限公司 | Resource information recommends method and resource information commending system |
| CN108694542A (en) * | 2018-05-14 | 2018-10-23 | 北京辰森世纪科技股份有限公司 | Processing method, device and the equipment of applying and purchasing the experiment materials information |
| CN108829819A (en) * | 2018-06-12 | 2018-11-16 | 上海智臻智能网络科技股份有限公司 | Personalized text recommended method and system, server, readable storage medium storing program for executing |
| CN108932643A (en) * | 2017-05-25 | 2018-12-04 | 苏宁云商集团股份有限公司 | A kind of personalized recommendation method and device |
| CN109389453A (en) * | 2017-08-11 | 2019-02-26 | 苏宁云商集团股份有限公司 | A kind of price analysis method and device |
| CN109903104A (en) * | 2017-12-08 | 2019-06-18 | 北京京东尚科信息技术有限公司 | Method and device for data analysis |
| CN109934604A (en) * | 2017-12-15 | 2019-06-25 | 北京京东尚科信息技术有限公司 | Obtain method, system, storage medium and the electronic equipment of best seller list |
| CN110494881A (en) * | 2017-03-29 | 2019-11-22 | 瑞可利有限公司 | For managing the system, method and program of message |
| CN110717097A (en) * | 2019-09-06 | 2020-01-21 | 中国平安财产保险股份有限公司 | Service recommendation method and device, computer equipment and storage medium |
| CN111199459A (en) * | 2019-12-30 | 2020-05-26 | 深圳市盟天科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
| CN111882409A (en) * | 2020-09-28 | 2020-11-03 | 武汉卓尔数字传媒科技有限公司 | Method and device for recommending main body and electronic equipment |
| CN113807926A (en) * | 2021-09-26 | 2021-12-17 | 北京沃东天骏信息技术有限公司 | Recommendation information generation method and device, electronic equipment and computer readable medium |
| CN114663200A (en) * | 2022-05-23 | 2022-06-24 | 中国平安财产保险股份有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| CN114708070A (en) * | 2022-06-06 | 2022-07-05 | 广东鑫兴科技有限公司 | Intelligent information pushing method and device, electronic equipment and storage medium |
| CN114781624A (en) * | 2022-05-23 | 2022-07-22 | 百色市恒赢电脑网络有限公司 | User behavior intention mining method based on big data analysis and big data system |
| CN111552883B (en) * | 2020-05-13 | 2023-12-12 | 咪咕文化科技有限公司 | Content recommendation method and computer-readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
| CN102479366A (en) * | 2010-11-25 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Commodity recommending method and system |
| CN102956009A (en) * | 2011-08-16 | 2013-03-06 | 阿里巴巴集团控股有限公司 | Electronic commerce information recommending method and electronic commerce information recommending device on basis of user behaviors |
| CN103700005A (en) * | 2013-12-17 | 2014-04-02 | 南京信息工程大学 | Association-rule recommending method based on self-adaptive multiple minimum supports |
-
2014
- 2014-07-16 CN CN201410338542.XA patent/CN105321089A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102479366A (en) * | 2010-11-25 | 2012-05-30 | 阿里巴巴集团控股有限公司 | Commodity recommending method and system |
| CN102956009A (en) * | 2011-08-16 | 2013-03-06 | 阿里巴巴集团控股有限公司 | Electronic commerce information recommending method and electronic commerce information recommending device on basis of user behaviors |
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
| CN103700005A (en) * | 2013-12-17 | 2014-04-02 | 南京信息工程大学 | Association-rule recommending method based on self-adaptive multiple minimum supports |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017143701A1 (en) * | 2016-02-22 | 2017-08-31 | 青岛海尔股份有限公司 | Method and device for processing information on the basis of food quantity |
| CN105761111A (en) * | 2016-02-22 | 2016-07-13 | 青岛海尔股份有限公司 | Information processing method and information processing device based on dosage of ingredients |
| CN107730336A (en) * | 2016-08-12 | 2018-02-23 | 苏宁云商集团股份有限公司 | Commodity method for pushing and device in a kind of online transaction |
| CN107730337A (en) * | 2016-08-12 | 2018-02-23 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN107093120A (en) * | 2016-09-28 | 2017-08-25 | 北京小度信息科技有限公司 | Service strategy method for pushing and device |
| CN107918882A (en) * | 2016-10-11 | 2018-04-17 | 阿里巴巴集团控股有限公司 | The method and device of data object information is provided |
| CN108009867B (en) * | 2016-10-28 | 2021-04-30 | 上海优扬新媒信息技术有限公司 | Information output method and device |
| CN108009867A (en) * | 2016-10-28 | 2018-05-08 | 百度在线网络技术（北京）有限公司 | Information output method and device |
| CN108205768A (en) * | 2016-12-20 | 2018-06-26 | 百度在线网络技术（北京）有限公司 | Database building method and data recommendation method and device, equipment and storage medium |
| CN108205768B (en) * | 2016-12-20 | 2022-07-05 | 百度在线网络技术（北京）有限公司 | Database establishing method, data recommending device, equipment and storage medium |
| CN106910108A (en) * | 2017-01-24 | 2017-06-30 | 武汉奇米网络科技有限公司 | A kind of items list methods of exhibiting and system |
| CN110494881A (en) * | 2017-03-29 | 2019-11-22 | 瑞可利有限公司 | For managing the system, method and program of message |
| CN108932643A (en) * | 2017-05-25 | 2018-12-04 | 苏宁云商集团股份有限公司 | A kind of personalized recommendation method and device |
| CN107358508A (en) * | 2017-07-18 | 2017-11-17 | 广州优视网络科技有限公司 | A kind of similar article item management method and device |
| CN107402742A (en) * | 2017-08-04 | 2017-11-28 | 北京京东尚科信息技术有限公司 | Information-pushing method and device |
| CN109389453A (en) * | 2017-08-11 | 2019-02-26 | 苏宁云商集团股份有限公司 | A kind of price analysis method and device |
| CN107657506A (en) * | 2017-09-06 | 2018-02-02 | 北京五八到家信息技术有限公司 | The intelligent recommendation method and commending system of a kind of service commodity |
| CN107679905A (en) * | 2017-09-28 | 2018-02-09 | 千寻位置网络有限公司 | Market guidance method and system, service terminal, memory based on customer location |
| CN109903104A (en) * | 2017-12-08 | 2019-06-18 | 北京京东尚科信息技术有限公司 | Method and device for data analysis |
| CN109934604B (en) * | 2017-12-15 | 2021-09-07 | 北京京东尚科信息技术有限公司 | Sales data processing method and system, storage medium and electronic equipment |
| CN109934604A (en) * | 2017-12-15 | 2019-06-25 | 北京京东尚科信息技术有限公司 | Obtain method, system, storage medium and the electronic equipment of best seller list |
| CN108446963A (en) * | 2018-03-30 | 2018-08-24 | 深圳春沐源控股有限公司 | Resource information recommends method and resource information commending system |
| WO2019184204A1 (en) * | 2018-03-30 | 2019-10-03 | 深圳春沐源控股有限公司 | Resource information recommendation method and resource information recommendation system |
| CN108694542A (en) * | 2018-05-14 | 2018-10-23 | 北京辰森世纪科技股份有限公司 | Processing method, device and the equipment of applying and purchasing the experiment materials information |
| CN108829819B (en) * | 2018-06-12 | 2021-05-07 | 上海智臻智能网络科技股份有限公司 | Personalized text recommendation method and system, server and readable storage medium |
| CN108829819A (en) * | 2018-06-12 | 2018-11-16 | 上海智臻智能网络科技股份有限公司 | Personalized text recommended method and system, server, readable storage medium storing program for executing |
| CN110717097A (en) * | 2019-09-06 | 2020-01-21 | 中国平安财产保险股份有限公司 | Service recommendation method and device, computer equipment and storage medium |
| CN111199459A (en) * | 2019-12-30 | 2020-05-26 | 深圳市盟天科技有限公司 | Commodity recommendation method and device, electronic equipment and storage medium |
| CN111199459B (en) * | 2019-12-30 | 2024-02-09 | 深圳市盟天科技有限公司 | Commodity recommendation method, commodity recommendation device, electronic equipment and storage medium |
| CN111552883B (en) * | 2020-05-13 | 2023-12-12 | 咪咕文化科技有限公司 | Content recommendation method and computer-readable storage medium |
| CN111882409A (en) * | 2020-09-28 | 2020-11-03 | 武汉卓尔数字传媒科技有限公司 | Method and device for recommending main body and electronic equipment |
| CN113807926A (en) * | 2021-09-26 | 2021-12-17 | 北京沃东天骏信息技术有限公司 | Recommendation information generation method and device, electronic equipment and computer readable medium |
| CN114663200A (en) * | 2022-05-23 | 2022-06-24 | 中国平安财产保险股份有限公司 | Product recommendation method and device, electronic equipment and storage medium |
| CN114781624A (en) * | 2022-05-23 | 2022-07-22 | 百色市恒赢电脑网络有限公司 | User behavior intention mining method based on big data analysis and big data system |
| CN114708070A (en) * | 2022-06-06 | 2022-07-05 | 广东鑫兴科技有限公司 | Intelligent information pushing method and device, electronic equipment and storage medium |
| CN114708070B (en) * | 2022-06-06 | 2022-09-23 | 广东鑫兴科技有限公司 | Intelligent information pushing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105321089A (en) | Method and system for e-commerce recommendation based on multi-algorithm fusion | |
| JP6356744B2 (en) | Method and system for displaying cross-website information | |
| CN109685631B (en) | Personalized recommendation method based on big data user behavior analysis | |
| CN109299370B (en) | Multi-pair level personalized recommendation method | |
| US10157344B2 (en) | Systems and methods for trend aware self-correcting entity relationship extraction | |
| US9400831B2 (en) | Providing information recommendations based on determined user groups | |
| JP5976943B2 (en) | Product information recommendation | |
| WO2021139164A1 (en) | Sequential recommendation method based on long-term interest and short-term interest | |
| TWI534734B (en) | Feed apparatus, information providing method, program, and recording medium | |
| CN102073717A (en) | Home page recommending method for orienting vertical e-commerce website | |
| US11727014B2 (en) | Dynamic filter recommendations | |
| CN105589905A (en) | User interest data analysis and collection system and method | |
| US20160063590A1 (en) | Dynamic taxonomy generation with demand-based product groups | |
| CN103886001A (en) | Personalized commodity recommendation system | |
| CN105528374A (en) | A commodity recommendation method in electronic commerce and a system using the same | |
| CN109658206A (en) | Information recommendation method and device | |
| CN104317945A (en) | E-commerce website commodity recommending method on basis of search behaviors | |
| JP2010134733A (en) | Information recommendation device, information recommendation method, and program | |
| TW201413619A (en) | Recommending product information | |
| CN106447463A (en) | Commodity recommendation method based on Markov decision-making process model | |
| CN102629360A (en) | Effective dynamic commodity recommendation method and commodity recommendation system | |
| CN107274242A (en) | A kind of Method of Commodity Recommendation based on association analysis algorithm | |
| US20210118020A1 (en) | Price-Based User Feedback System | |
| CN113689259A (en) | Commodity personalized recommendation method and system based on user behaviors | |
| CN108596712B (en) | Single-class collaborative filtering method based on article sequence, storage medium and server |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160210 |
|
| RJ01 | Rejection of invention patent application after publication |