CN101783004A - Fast intelligent commodity recommendation system - Google Patents
Fast intelligent commodity recommendation system Download PDFInfo
- Publication number
- CN101783004A CN101783004A CN201010119583A CN201010119583A CN101783004A CN 101783004 A CN101783004 A CN 101783004A CN 201010119583 A CN201010119583 A CN 201010119583A CN 201010119583 A CN201010119583 A CN 201010119583A CN 101783004 A CN101783004 A CN 101783004A
- Authority
- CN
- China
- Prior art keywords
- commodity
- businessman
- client
- commercial product
- recommended
- 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
Images
Abstract
The invention discloses a fast intelligent commodity recommendation system. The recommendation system provides personalized services for customers according to own demands of a commodity provider. The recommendation system comprises a merchant demand analysis module and a commodity recommendation module. A merchant demand is expressed in a form of rules; and a customer transaction database stores transaction records of a merchant. The merchant demand analysis module evaluates commodities operated by the merchant according to the merchant demand and the customer transaction database to generate a commodity library to be recommended. The commodity recommendation module generates a recommended commodity list meeting the own requirement of a current customer according to the customer transaction database and the commodity library to be recommended to provide commodity services for the current customer. The commodity recommendation generated by the system is not only customized for the customer, but also is customized for the merchant, can greatly increase the sales and the profits for the merchant, can improve the service speed of the commodity recommendation system to realize real-time service, has high expandability, and can meet the requirements of large merchants.
Description
Technical field
The present invention relates to commercial product recommending system, be specifically related to the commercial product recommending system of fast intelligent.
Background technology
Commodity comprise products ﹠ services.Commercial product recommending comprises advertisement and preferential plan, is the core means of businessman's sales promotion.Traditional commercial product recommending right and wrong personalization, can not satisfy individual demand.It is a focus of present global businesses that personalized commercial product recommending is provided, and can improve the effect of commercial product recommending greatly, the Amazon of the U.S. for example, and after using the individual commodity commending system in 2003, annual sales volume has improved 25%.The large supermarket of the U.S. or online sales merchant, as Wal-Mart, Target, CVS, Amazon all attaches great importance to personalized commercial product recommending.Domestic present personalized commercial product recommending is also not general.There is following defective in existing personalized commercial product recommending system:
1, the demand of having ignored businessman, real intelligentized commercial product recommending system not only will be considered client's demands of individuals, also will consider the demand of businessman, such as some best seller, even do not carry out advertising, also can sell very soon;
2, can not handle mass data, there are tens thousand of kinds of commodity in large-scale businessman, hundreds thousand of times of the transaction of every day, accumulated the transaction data of magnanimity, these transaction data have very great help to intelligentized commercial product recommending is provided, but how to handle such mass data flow, it is a very problem of difficulty, if can not solve, the response time of commending system can be very slow, and system can't be practical, perhaps only can be used for little businessman, and can not be used in large-scale department store and the supermarket chain;
3, slow to the reaction of turn of the market, many commercial product recommending systems upgrade slowly, and a week or one month be new model, lacks update method fast.
Summary of the invention
The objective of the invention is to: the commercial product recommending system that a kind of fast intelligent is provided, this commending system can be according to commodity provider self needs, for the client provides personalized commercial product recommending service, and can the fast processing mass data, for use in large-scale businessman.
Technical solution of the present invention is: this commending system comprises businessman's demand analysis module and commercial product recommending module, businessman's demand is expressed with rule format, the client trading database stores transaction record of businessman, businessman's demand analysis module is assessed the commodity that businessman manages according to businessman's demand and client trading database, produce commodity to be recommended storehouse, the commercial product recommending module produces the Recommendations tabulation of satisfying current client's demands of individuals according to client trading database and commodity to be recommended storehouse and provides commodity and service for current client.
Wherein, in order to improve the service rate of commercial product recommending system, the commercial product recommending module has adopted two kinds of accelerated modes: (1) adopts off-line type to calculate, and carries out at non operation time usually; (2) adopt Distributed Calculation.
The invention has the advantages that:
1, the commercial product recommending that intellectuality, native system produce is not only to customization, also is businessman's customization, and this service targetedly can greatly increase the sales volume and the profit of businessman;
2, quick and high divergence, native system adopt the Distributed Calculation of accelerated mode or off-line type to calculate, and improve the service rate of commercial product recommending system, realize service in real time, have good expandability simultaneously, can satisfy the demand of large-scale businessman;
3, turn of the market is quick on the draw, native system adopts on-line study, and according to the business data on the turn of the market and the same day, the non operation time in the day of trade upgrades system, and self-adaptation is strong.
Description of drawings
Fig. 1 is a block diagram of the present invention.
Fig. 2 is the evaluation table that Fig. 1 generates according to transaction data.
Fig. 3 learns signal distributed the gathering of the similarity of Fig. 1.
Among the figure: 110 businessman's demands, 120 client trading databases, 130 businessmans demand analysis module, 140 commodity to be recommended storehouses, 150 commercial product recommending modules, 160 current clients.
Embodiment
As shown in Figure 1, this commending system comprises businessman's demand analysis module 130 and commercial product recommending module 150, businessman's demand 110 is expressed with rule format, client trading database 120 has stored the transaction record of businessman, businessman's demand analysis module 130 is assessed according to the commodity of businessman's demand 110 and 120 pairs of businessmans' operation of client trading database, produce commodity to be recommended storehouse 140, commercial product recommending module 150 produces the Recommendations tabulation of satisfying current client's demands of individuals according to client trading database 120 and commodity to be recommended storehouse 140 and provides commodity and service for current client.
Commercial product recommending involved in the present invention comprises advertisement and commercial promotions, and send mode can be papery, mobile phone or computer.
One group of regular expression of businessman's demand 110 usefulness, descriptive language are single order prediction logic (first order predictive logic), and perhaps other determines formula or probabilistic type rule description language, to express complex rule; The example of a single order prediction logic rule: commodity A is a new product, and belongs to washing class, then keypoint recommendation; This regular expression is: washing of A ∈ new product AND A ∈ class → keypoint recommendation A; Calculate a commercial product recommending index, the computing formula of this index is
Here R1 (X) is the recommendation index of commodity X, N
c(X) the rule sum of Recommendations X is supported in representative, and N (X) relates to the rule sum of commodity X, and R1 is an index between [0,1], and the big more expression of value businessman needs to recommend this commodity more.
Client trading database 120 has been stored the transaction data of businessman, this database is a relationship type, OO, or the relationship object type, this database can inquire the record of client, commodity, exchange hour fast, as a rule this database storing the transaction record of magnanimity, belong to large database.
The operation of businessman's demand analysis module 130 is as follows:
1, analyze client trading database 120, set up a probabilistic forecasting model, this forecast model is learnt from database, rather than expert's generation, be surprising because analyze the needed time of such lot of data by hand; The target variable R2 of this model expresses the probability of recommending certain commodity, and the value of R2 is between [0,1], and the big more expression of value businessman needs to recommend this commodity more; Businessman's demand analysis module 130 certain commodity of statistics are in the sales situation of certain time block, if commodity are unsalable more, R2 is big more; The predictive variable of forecast model (predictivevariable) has comprised economic environment, the sales season of market condition, and the position at businessman place, demographic situation, traffic etc.; The functional form of forecast model can be the decision tree class, discriminant analysis, Bayes classifier, neural network, lump study, the perhaps combination in any of these algorithms; After having produced such forecast model, can give a mark to each commodity, big more expression businessman needs to recommend this commodity more;
2, information fusion has two information sources now, and one is R2, another is that the recommendation index R (X) of final goods X is that the Bayes of R1 and R2 is average according to businessman's demand 110 rule R1 that produce, and the value of R is [0,1] between, if R＞0.5, we join X in the commodity to be recommended storehouse 140.
The functional block of commercial product recommending module 150 has two, the one, produce recommended models, the 2nd, be that current client produces Recommendations with recommended models, recommended models is based on collaborative filtering, be to utilize client's behavioral data in the client trading database 120 to predict client's preference, because client trading database 120 is high-volume databases, adopt based on the capable fast processing mass data of the collaborative filtering of commodity; Collaborative filtering based on commodity is to be operated on the evaluation table (rating table), and evaluation table produces from client trading database 120, as shown in Figure 2, the capable representative of consumer of evaluation table, row are represented commodity; Calculate similarity between the commodity according to evaluation table; To commodity X
iAnd X
j, the similarity S (X between them
i, X
j) be used to portray the possibility that the client buys two commodity simultaneously, similarity S related coefficient class, the cosine coefficient class, perhaps other is out of shape and expresses; S is the core of recommended models, given S, and when serving a current client, commending system utilizes method of weighted mean to commodity i calculated recommendation degree P
U, j,
Here y
U, jBe the evaluation of client u to commodity j, S
I, nBe the similarity between commodity i and the n, N is the commodity set similar to commodity i; System selects P
U, jA highest 1-5 commodity are recommended current client.
Described the production method of the quick high divergence of S here, the calculating of S only considers to have estimated simultaneously X
iAnd X
jThe client, the part that marks with black matrix among Fig. 2, the set that these clients form represents that with U cosine similarity S computing formula is as follows
Here y
U, jBe the evaluation of client u to commodity j, notice that cosine similarity S has two characteristics: (1) S expresses with fully adding up (sufficient statistics), and in formula 3, the quantity of information of data is expressed as three and formula, store the value of three and formula, just determined S fully; (2) each all is decomposable to the client with formula in the formula 3, and the client is divided into different subclass, each subclass is carried out computing, and then gather and calculate.
At these characteristics of similarity, the present invention proposes the distributed learning method that gathers of similarity, Fig. 3 shows the overall design of algorithm, and algorithm is used for any quick calculating that fully statistics and client are divided the similarity of concrete conditions in the establishment of a specific crime of satisfying, and is achieved as follows:
A) client is collected U and be divided into several subclass { U
k(1≤k≤K); Each subclass U
kData handle by a computing node, each node carries out this locality and calculates, and draws the Z1=∑
U ∈ Uky
U, iy
U, j, the Z2=∑
U ∈ Uky
U, iy
U, i, the Z3=∑
U ∈ Uiy
U, jy
U, j.
B) each node is with Z1, and Z2, Z3 deliver to Centroid, intersects study, obtains similarities according to formula 3.
The distributed learning method that gathers makes native system that fabulous divergence be arranged, and can come linear shortening computing time by increasing the node number; The present invention obtains the ability of fast processing mass data in the following manner, (1) distributed shortening computing time of gathering the learning method linearity; (2) commercial product recommending is only considered in the commodity to be recommended storehouse 140 commodity, has significantly reduced calculated amount; (3) calculation of similarity degree is an off-line, can finish at non operation time.
The update method of system is described below: system upgrades the accuracy that guarantees system very important according to the variation and the new client trading data of market condition; When market condition changes, system will upgrade businessman's demand 110; According to new transaction data, system update probabilistic forecasting model has been finished the renewal of businessman's demand analysis module 130 so simultaneously; The renewal of commercial product recommending module 150 mainly is the renewal of similar kilsyth basalt, utilizes the decomposability and the abundant statistical of similarity, new customer data U
NewRepresent that system calculates Z1 earlier
New=∑
U ∈ Unewy
U, iy
U, j, Z2
New=∑
U ∈ Unewy
U, iy
U, i, Z3
New=∑
U ∈ Unewy
U, jy
U, j. then with Z1
New, Z2
New, Z3
NewBring the similarity that formula 3 acquisitions are upgraded into.
Implementation method of the present invention has above been described.The present invention can be used for many application.Although the description of being done is related with specific realization example, the invention thought of core and technological treatment are applicable to other application.Modification and the distortion made on core concept of the present invention and technological treatment all belong to protection scope of the present invention.
Claims (6)
1. the commercial product recommending system of fast intelligent, this commending system is according to commodity provider self needs, for the client provides personalized service, it is characterized in that: this commending system comprises businessman's demand analysis module (130) and commercial product recommending module (150), businessman's demand (110) is expressed with rule format, client trading database (120) has stored the transaction record of businessman, businessman's demand analysis module (130) is assessed the commodity that businessman manages according to businessman's demand (110) and client trading database (120), produce commodity to be recommended storehouse (140), commercial product recommending module (150) produces the Recommendations tabulation of satisfying current client's demands of individuals according to client trading database (120) and commodity to be recommended storehouse (140) and provides commodity and service for current client.
2. the commercial product recommending system of fast intelligent according to claim 1, it is characterized in that: this system produces the commodity to be recommended storehouse based on businessman's demand and sales data, produce one according to businessman's demand and recommend index, produce another according to sales data and recommend index, it is the average of two indexes that final goods is recommended index.
3. the commercial product recommending system of fast intelligent according to claim 1, it is characterized in that: businessman's demand is with one group of regular expression, descriptive language is to determine formula or probabilistic type rule description language, to express complex rule, calculates a commercial product recommending index according to these rules.
4. the commercial product recommending system of fast intelligent according to claim 1, it is characterized in that: businessman's demand analysis module is added up the sales situation of certain commodity in certain time period, set up a forecast model, the predictive variable of forecast model has comprised the market condition economic environment, sales season, and the position at businessman place, demographic situation and traffic, the functional form of forecast model is single arbitrarily sorter, the perhaps combination in any of these sorters, after having produced such forecast model, to the marking of each commodity, big more expression businessman needs to recommend this commodity more.
5. the commercial product recommending system of fast intelligent according to claim 1, it is characterized in that: the commercial product recommending module produces recommended models fast, and recommended models is to be operated on the evaluation table, and evaluation table produces from the client trading database, the capable representative of consumer of evaluation table, row are represented commodity; Calculate similarity between the commodity according to evaluation table, to commodity X
iAnd X
j, the similarity S (X between them
i, X
j) be used to portray the possibility that the client buys two commodity simultaneously; S is the core of recommended models, given S, and when serving a current client, commending system utilizes method of weighted mean to commodity calculated recommendation degree; Satisfy the quick calculating that fully statistics and client can be divided the similarity of concrete conditions in the establishment of a specific crime for any, the producing method of the quick high divergence of S is as follows:
A. the client is collected U and be divided into several subclass { U
k(1≤k≤K);
B. each subclass U
kData handle by a computing node, each node carries out this locality and calculates, and uses y
U, jBe the evaluation of client u, based on local data { y to commodity j
U, j(u ∈ U
k), produce one group of local sufficient statistic;
C. each computing node is delivered to a Centroid with local sufficient statistic, intersects study, calculates the sufficient statistic that gathers, and according to the sufficient statistic that gathers, calculates similarity.
6. the commercial product recommending system of fast intelligent according to claim 5, it is characterized in that: this commending system comprises the renewal of commodity to be recommended storehouse and recommended models; When market condition changes, system will upgrade businessman's demand; Simultaneously according to new transaction data, system update probabilistic forecasting model; Finished after these the two groups renewals, the recommendation index of commodity changes, according to new recommendation index update commodity to be recommended storehouse; The renewal of similar kilsyth basalt has been based on the decomposability and the abundant statistical of similarity, to new transaction data, calculates fully statistics, this group is fully added up being updated in original abundant statistics, upgrades.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010119583A CN101783004A (en) | 2010-03-03 | 2010-03-03 | Fast intelligent commodity recommendation system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010119583A CN101783004A (en) | 2010-03-03 | 2010-03-03 | Fast intelligent commodity recommendation system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101783004A true CN101783004A (en) | 2010-07-21 |
Family
ID=42522990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010119583A Pending CN101783004A (en) | 2010-03-03 | 2010-03-03 | Fast intelligent commodity recommendation system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101783004A (en) |
Cited By (31)
| 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 |
| CN102592223A (en) * | 2011-01-18 | 2012-07-18 | 卓望数码技术（深圳）有限公司 | Commodity recommending method and commodity recommending system |
| CN102750274A (en) * | 2011-01-04 | 2012-10-24 | 张越峰 | Field intelligent guide service system and method with preliminary human thinking |
| CN103020855A (en) * | 2012-12-11 | 2013-04-03 | 北京京东世纪贸易有限公司 | Bad commodity distinguishing method and system based on user purchasing behavior |
| CN103294812A (en) * | 2013-06-06 | 2013-09-11 | 浙江大学 | Commodity recommendation method based on mixed model |
| CN103345699A (en) * | 2013-07-10 | 2013-10-09 | 湖南大学 | Personalized food recommendation method based on commodity forest system |
| CN103886489A (en) * | 2014-04-16 | 2014-06-25 | 深圳汇费通科技有限公司 | Method achieving accurate promotion information delivery according to cloud data |
| CN104537100A (en) * | 2011-01-04 | 2015-04-22 | 张越峰 | Intelligence system and method with primary human thinking |
| CN104657487A (en) * | 2015-03-05 | 2015-05-27 | 东方网力科技股份有限公司 | Licence plate recommendation method and device based on user licence plate querying behavior |
| CN105009156A (en) * | 2012-12-20 | 2015-10-28 | 沃尔玛连锁商店公司 | Framework for generating a personalized item list |
| CN105069633A (en) * | 2015-08-06 | 2015-11-18 | 北京优凯澜信息技术有限公司 | Artwork O2O platform implementation method |
| CN105069651A (en) * | 2015-08-06 | 2015-11-18 | 北京优凯澜信息技术有限公司 | Artwork O2O platform implementation method |
| CN105761122A (en) * | 2016-04-29 | 2016-07-13 | 山东大学 | Product recommendation method and device fusing manufacturer similarity |
| CN106033576A (en) * | 2015-03-11 | 2016-10-19 | 哈尔滨工业大学深圳研究生院 | Method and apparatus for displaying object information |
| WO2017000185A1 (en) * | 2015-06-30 | 2017-01-05 | 深圳市星电商科技有限公司 | Data exchange processing method, terminal, and system |
| CN107239879A (en) * | 2017-05-02 | 2017-10-10 | 太仓鸿策拓达科技咨询有限公司 | A kind of service push method for being used to facilitate financial transaction |
| CN107357888A (en) * | 2017-07-10 | 2017-11-17 | 北京小度信息科技有限公司 | The offer method, apparatus and electronic equipment of raw material information |
| CN107403345A (en) * | 2017-09-22 | 2017-11-28 | 北京京东尚科信息技术有限公司 | Best-selling product Forecasting Methodology and system, storage medium and electric terminal |
| CN103839172B (en) * | 2012-11-23 | 2017-12-29 | 阿里巴巴集团控股有限公司 | Method of Commodity Recommendation and system |
| CN108320171A (en) * | 2017-01-17 | 2018-07-24 | 北京京东尚科信息技术有限公司 | Hot item prediction technique, system and device |
| CN108932648A (en) * | 2017-07-24 | 2018-12-04 | 上海宏原信息科技有限公司 | A kind of method and apparatus for predicting its model of item property data and training |
| CN109120658A (en) * | 2017-06-23 | 2019-01-01 | 杭州美界科技有限公司 | A kind of merchant end and the parallel beauty recommender system of background server |
| CN110363558A (en) * | 2018-04-11 | 2019-10-22 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus generating commodity association message |
| CN110998619A (en) * | 2017-07-28 | 2020-04-10 | 纽诺有限公司 | System and mechanism for marketing products on autonomous vehicles |
| CN111127077A (en) * | 2019-11-29 | 2020-05-08 | 中国建设银行股份有限公司 | Recommendation method and device based on stream computing |
| CN111507486A (en) * | 2019-01-29 | 2020-08-07 | 买风格科技股份有限公司 | Environment-friendly intelligent circulation retail method and system for supplying and selling various life single products in series |
| CN111553715A (en) * | 2020-04-29 | 2020-08-18 | 惠州视维新技术有限公司 | Information feedback method, server and readable storage medium |
| CN112001777A (en) * | 2020-08-24 | 2020-11-27 | 知小二(广州)科技有限公司 | E-commerce platform data analysis system based on cloud computing |
| CN112541111A (en) * | 2020-11-09 | 2021-03-23 | 武汉蝌蚪信息技术有限公司 | Commodity retrieval and commodity recommendation system based on decentralized big data retrieval market |
| CN113469719A (en) * | 2020-03-30 | 2021-10-01 | 青岛海尔电冰箱有限公司 | Sales floor refrigerator recommendation method, sales floor navigation device and computer readable storage medium |
| CN113469718A (en) * | 2020-03-30 | 2021-10-01 | 青岛海尔电冰箱有限公司 | Sales floor refrigerator recommendation method, sales floor navigation device and computer readable storage medium |
-
2010
- 2010-03-03 CN CN201010119583A patent/CN101783004A/en active Pending
Cited By (42)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102750274A (en) * | 2011-01-04 | 2012-10-24 | 张越峰 | Field intelligent guide service system and method with preliminary human thinking |
| CN104537100B (en) * | 2011-01-04 | 2019-03-19 | 张越峰 | A kind of intelligence system and method for just tool human thinking |
| CN104537100A (en) * | 2011-01-04 | 2015-04-22 | 张越峰 | Intelligence system and method with primary human thinking |
| CN102592223A (en) * | 2011-01-18 | 2012-07-18 | 卓望数码技术（深圳）有限公司 | Commodity recommending method and commodity recommending system |
| CN102411754A (en) * | 2011-11-29 | 2012-04-11 | 南京大学 | Personalized recommendation method based on commodity property entropy |
| CN103839172B (en) * | 2012-11-23 | 2017-12-29 | 阿里巴巴集团控股有限公司 | Method of Commodity Recommendation and system |
| CN103020855A (en) * | 2012-12-11 | 2013-04-03 | 北京京东世纪贸易有限公司 | Bad commodity distinguishing method and system based on user purchasing behavior |
| CN103020855B (en) * | 2012-12-11 | 2016-02-10 | 北京京东世纪贸易有限公司 | The method and system of bad commodity is identified based on user's buying behavior |
| CN105009156A (en) * | 2012-12-20 | 2015-10-28 | 沃尔玛连锁商店公司 | Framework for generating a personalized item list |
| CN103294812B (en) * | 2013-06-06 | 2016-02-10 | 浙江大学 | A kind of Method of Commodity Recommendation based on mixture model |
| CN103294812A (en) * | 2013-06-06 | 2013-09-11 | 浙江大学 | Commodity recommendation method based on mixed model |
| CN103345699A (en) * | 2013-07-10 | 2013-10-09 | 湖南大学 | Personalized food recommendation method based on commodity forest system |
| CN103886489A (en) * | 2014-04-16 | 2014-06-25 | 深圳汇费通科技有限公司 | Method achieving accurate promotion information delivery according to cloud data |
| CN104657487B (en) * | 2015-03-05 | 2017-12-22 | 东方网力科技股份有限公司 | A kind of car plate based on user's car plate User behavior recommends method and device |
| CN104657487A (en) * | 2015-03-05 | 2015-05-27 | 东方网力科技股份有限公司 | Licence plate recommendation method and device based on user licence plate querying behavior |
| CN106033576B (en) * | 2015-03-11 | 2020-08-11 | 哈尔滨工业大学深圳研究生院 | Object information display method and device |
| CN106033576A (en) * | 2015-03-11 | 2016-10-19 | 哈尔滨工业大学深圳研究生院 | Method and apparatus for displaying object information |
| WO2017000185A1 (en) * | 2015-06-30 | 2017-01-05 | 深圳市星电商科技有限公司 | Data exchange processing method, terminal, and system |
| CN105069633A (en) * | 2015-08-06 | 2015-11-18 | 北京优凯澜信息技术有限公司 | Artwork O2O platform implementation method |
| CN105069651A (en) * | 2015-08-06 | 2015-11-18 | 北京优凯澜信息技术有限公司 | Artwork O2O platform implementation method |
| CN105761122B (en) * | 2016-04-29 | 2020-09-08 | 山东大学 | Product recommendation method and device fusing manufacturer similarity |
| CN105761122A (en) * | 2016-04-29 | 2016-07-13 | 山东大学 | Product recommendation method and device fusing manufacturer similarity |
| CN108320171A (en) * | 2017-01-17 | 2018-07-24 | 北京京东尚科信息技术有限公司 | Hot item prediction technique, system and device |
| CN108320171B (en) * | 2017-01-17 | 2020-11-03 | 北京京东尚科信息技术有限公司 | Hot-sold commodity prediction method, system and device |
| CN107239879A (en) * | 2017-05-02 | 2017-10-10 | 太仓鸿策拓达科技咨询有限公司 | A kind of service push method for being used to facilitate financial transaction |
| CN109120658A (en) * | 2017-06-23 | 2019-01-01 | 杭州美界科技有限公司 | A kind of merchant end and the parallel beauty recommender system of background server |
| CN107357888B (en) * | 2017-07-10 | 2021-06-15 | 北京星选科技有限公司 | Method and device for providing raw material information and electronic equipment |
| CN107357888A (en) * | 2017-07-10 | 2017-11-17 | 北京小度信息科技有限公司 | The offer method, apparatus and electronic equipment of raw material information |
| CN108932648A (en) * | 2017-07-24 | 2018-12-04 | 上海宏原信息科技有限公司 | A kind of method and apparatus for predicting its model of item property data and training |
| CN110998619B (en) * | 2017-07-28 | 2023-10-24 | 纽诺有限公司 | System and mechanism for adding and selling products on autonomous carriers |
| US11830058B2 (en) | 2017-07-28 | 2023-11-28 | Nuro, Inc. | Automated retail store on autonomous or semi-autonomous vehicle |
| CN110998619A (en) * | 2017-07-28 | 2020-04-10 | 纽诺有限公司 | System and mechanism for marketing products on autonomous vehicles |
| CN107403345A (en) * | 2017-09-22 | 2017-11-28 | 北京京东尚科信息技术有限公司 | Best-selling product Forecasting Methodology and system, storage medium and electric terminal |
| CN110363558A (en) * | 2018-04-11 | 2019-10-22 | 北京京东尚科信息技术有限公司 | A kind of method and apparatus generating commodity association message |
| CN111507486A (en) * | 2019-01-29 | 2020-08-07 | 买风格科技股份有限公司 | Environment-friendly intelligent circulation retail method and system for supplying and selling various life single products in series |
| CN111127077A (en) * | 2019-11-29 | 2020-05-08 | 中国建设银行股份有限公司 | Recommendation method and device based on stream computing |
| CN113469719A (en) * | 2020-03-30 | 2021-10-01 | 青岛海尔电冰箱有限公司 | Sales floor refrigerator recommendation method, sales floor navigation device and computer readable storage medium |
| CN113469718A (en) * | 2020-03-30 | 2021-10-01 | 青岛海尔电冰箱有限公司 | Sales floor refrigerator recommendation method, sales floor navigation device and computer readable storage medium |
| CN111553715A (en) * | 2020-04-29 | 2020-08-18 | 惠州视维新技术有限公司 | Information feedback method, server and readable storage medium |
| CN112001777A (en) * | 2020-08-24 | 2020-11-27 | 知小二(广州)科技有限公司 | E-commerce platform data analysis system based on cloud computing |
| CN112001777B (en) * | 2020-08-24 | 2023-12-08 | 珠海以备科技有限责任公司 | Electronic commerce platform data analysis system based on cloud computing |
| CN112541111A (en) * | 2020-11-09 | 2021-03-23 | 武汉蝌蚪信息技术有限公司 | Commodity retrieval and commodity recommendation system based on decentralized big data retrieval market |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101783004A (en) | Fast intelligent commodity recommendation system | |
| Ren et al. | Demand forecasting in retail operations for fashionable products: methods, practices, and real case study | |
| Ni et al. | A two-stage dynamic sales forecasting model for the fashion retail | |
| US20200090195A1 (en) | Electronic neural network system for dynamically producing predictive data using varying data | |
| Han et al. | Segmentation of telecom customers based on customer value by decision tree model | |
| Luo | Analyzing the impact of social networks and social behavior on electronic business during COVID-19 pandemic | |
| CN110392899A (en) | The dynamic feature selection generated for model | |
| Cheng et al. | Customer lifetime value prediction by a Markov chain based data mining model: Application to an auto repair and maintenance company in Taiwan | |
| CN111652657A (en) | Commodity sales prediction method and device, electronic equipment and readable storage medium | |
| CN102272758A (en) | Automated specification, estimation, discovery of causal drivers and market response elasticities or lift factors | |
| CN113553540A (en) | Commodity sales prediction method | |
| JPWO2018207259A1 (en) | Information processing system, information processing apparatus, prediction model extraction method, and prediction model extraction program | |
| Rao et al. | Research on personalized referral service and big data mining for e-commerce with machine learning | |
| CN111582926A (en) | High-end member management system | |
| JP2011134356A (en) | Campaign dynamic correction system, method thereof, recording medium containing the method, and transmission medium for transmitting the method | |
| CN115409577A (en) | Intelligent container repurchase prediction method and system based on user behavior and environmental information | |
| CN114372848A (en) | Tobacco industry intelligent marketing system based on machine learning | |
| Guo et al. | A deep prediction network for understanding advertiser intent and satisfaction | |
| Xu et al. | Optimal decision of multiobjective and multiperiod anticipatory shipping under uncertain demand: A data-driven framework | |
| Chen et al. | Hierarchical demand forecasting for factory production of perishable goods | |
| Xia et al. | The research of online shopping customer churn prediction based on integrated learning | |
| Zhou et al. | The allocation optimization of promotion budget and traffic volume for an online flash-sales platform | |
| CN116304374B (en) | Customer matching method and system based on package data | |
| Sun et al. | Deep learning based customer preferences analysis in Industry 4.0 Environment | |
| Durango-Cohen | Modeling contribution behavior in fundraising: Segmentation analysis for a public broadcasting station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100721 |