CN115392406A - User classification management method based on historical transaction information - Google Patents
User classification management method based on historical transaction information Download PDFInfo
- Publication number
- CN115392406A CN115392406A CN202211330543.0A CN202211330543A CN115392406A CN 115392406 A CN115392406 A CN 115392406A CN 202211330543 A CN202211330543 A CN 202211330543A CN 115392406 A CN115392406 A CN 115392406A
- Authority
- CN
- China
- Prior art keywords
- user
- undirected graph
- users
- transaction information
- historical transaction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
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]
-
- 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
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Technology Law (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the field of data processing, in particular to a user classification management method based on historical transaction information. The method includes the steps of obtaining historical transaction information of each user, constructing a first completely undirected graph of each user according to the historical transaction information of each user, trimming the first undirected graph to obtain a second completely undirected graph, obtaining the same parts of two second completely undirected graphs of any two users as sub-graph pairs of the two users, obtaining the degree of membership between the users according to the sub-graph pairs of the two users, constructing a completely undirected graph of user consumption relations according to the degree of membership between the users, obtaining a plurality of user types according to the completely undirected graph of the user consumption relations, and effectively improving accuracy of user classification according to consumption habit characteristics of the users.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a user classification management method based on historical transaction information.
Background
Through user classification of historical transaction information of the e-commerce platform users, the accuracy of recommending commodities to the users can be improved, and the commodity transaction amount of the e-commerce platform is increased.
However, in the existing classification method for the electric business users, when the users are classified based on the historical transaction information of the users, the electric business users are generally classified by adopting a consumption information matching method, but due to the diversity of consumption habits of different users, the consumption habits of the users cannot be accurately and effectively distinguished when the consumption information is matched by using the conventional classification method.
According to the scheme, the user classification management method based on the historical transaction information is further provided, the degree of membership among different users is obtained by constructing graph structure data of the users, optimization of the method is completed, the consumption habits of the users can be distinguished, and more accurate user classification results based on the historical transaction information are completed.
Disclosure of Invention
In order to solve the above technical problems, an object of the present invention is to provide a user classification management method based on historical transaction information, which adopts the following technical solutions:
a method for user classification management based on historical transaction information, the method comprising:
acquiring historical transaction information of each user, wherein the historical transaction information of each user comprises the type of commodities purchased by each user and the purchase time of each type of commodities, and obtaining a first complete undirected graph of each user according to the historical transaction information of each user;
building the first completely undirected graph of each user to obtain a second completely undirected graph of each user, and obtaining a user set with various consumption habits of each user according to the second completely undirected graphs of all the users; obtaining the degree of membership among the users according to the user set with various consumption habits of each user;
and constructing a user consumption relation completely undirected graph according to the degree of membership among the users, and obtaining a plurality of user types according to the user consumption relation completely undirected graph.
Preferably, the formula for obtaining the first completely undirected graph of each user according to the historical transaction information of each user includes:
acquiring commodity type sets purchased by all users from all user historical transaction information sets, acquiring the purchase time of each user for each type of commodity, randomly combining any two types of commodities in the commodity type sets to obtain a commodity type pair, acquiring an interval time value set of the commodity type pair according to the purchase time of each user for the two types of commodities in each commodity type pair, wherein the interval time value set comprises a plurality of interval time values, counting the plurality of interval time values of the commodity type pair to obtain a purchase interval time histogram, acquiring a time interval value average value of the commodity type pair according to the purchase interval time histogram, and acquiring the total number of the interval time values in the interval time value set as a first number of the commodity type pair; forming a two-dimensional vector by the first quantity of the commodity type pairs and the average value of the time interval values, and recording the two-dimensional vector as a weight vector;
and constructing a first completely undirected graph by taking each type of commodity as a node and taking the weight vector as a variable weight.
Preferably, the method for obtaining the degree of affiliation between users according to the user set with various consumption habits of each user comprises:
splitting the second completely undirected graph of the user and the same sub-graph in the second completely undirected graph of each diversified consumption habit user to obtain a same sub-graph set, forming a sub-graph pair by the same sub-graph in the same sub-graph set to obtain a plurality of sub-graph pairs, subtracting edge weights of corresponding edges of two sub-graphs in each sub-graph pair to obtain a difference vector, taking a product value obtained by multiplying values of two dimensions in the difference vector as a comprehensive difference of each corresponding edge of the sub-graph pair, performing negative correlation mapping on the comprehensive difference to obtain a matching value of each corresponding edge of the sub-graph pair, averaging the matching values of all corresponding edges of the sub-graph pair to obtain a first matching value of each sub-graph pair, and taking the average value of the first matching values of all sub-graph pairs in the same sub-graph set as the degree of membership between the user and each diversified consumption habit user, namely the degree of membership between users.
Preferably, the method for constructing the first completely undirected graph of each user to obtain the second completely undirected graph of each user includes:
deleting the edge of the first edge in the first completely undirected graph of the user to obtain a middle completely undirected graph, wherein the first edge is the edge of which one element of the edge weight in the undirected graph is 0, and deleting the isolated node in the middle completely undirected graph to obtain a second completely undirected graph.
Preferably, the method for obtaining a user set with various consumption habits of each user according to the second completely undirected graph of all users comprises:
the number of edges of the second completely undirected graph of each user is obtained, a first user sequence is obtained by arranging all users from small to large according to the number of the edges of the second completely undirected graph of all the users, a set formed by all the users of which the position sequence of each user is greater than that of the user is marked as a user set with various consumption habits of each user, and the user set with various consumption habits of each user is formed by a plurality of users with various consumption habits.
Preferably, the formula for constructing the completely undirected graph of the user consumption relationship according to the degree of membership between users comprises:
and taking each user as a node, and taking the degree of membership between the users as an edge weight to construct a complete undirected graph, and recording the completely undirected graph as a user consumption relationship.
Preferably, the method for obtaining multiple user types according to the completely undirected graph of the user consumption relationship includes:
and segmenting the user consumption relation completely undirected graph by utilizing the maximum stream to obtain a plurality of segmentation blocks, and taking users corresponding to all nodes of each segmentation block as a user type to obtain a plurality of user types.
The technical scheme of the invention has the following beneficial effects: according to the historical transaction information of the users, the shopping habit diagram structure data of different users are obtained to represent, and then the subordinate degree of the different users in the consumption habits is obtained according to the shopping habit diagram structure data of the different users, so that the classification accuracy in the user classification according to the consumption habits of the users is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a user classification management method based on historical transaction information according to an embodiment of the present invention.
Detailed Description
To further illustrate the technical means and effects of the present invention for achieving the predetermined objects, the following detailed description of the user classification management method based on historical transaction information, its specific implementation, structure, features and effects will be given below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the user classification management method based on historical transaction information in detail with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a user classification management method based on historical transaction information according to an embodiment of the present invention is shown, where the method includes:
step S001: and acquiring historical transaction information of each user.
The e-commerce platform can obtain the purchase time of each user at different time points for different commodity types, and collects the purchase historical transaction information of each user, wherein the historical transaction information of each user comprises the commodity type purchased by each user and the purchase time of each type of commodity.
Step S002: and obtaining a first complete undirected graph of each user according to the historical transaction information of each user.
The method comprises the following specific steps:
establishing a first completely undirected graph by taking a single type commodity as a node for representing historical transaction information of a current user, wherein the calculation method of the edge weight of the first completely undirected graph is as follows:
the edge weight values of the two nodes are used for representing the characteristics of the user for purchasing the commodities corresponding to the two nodes at the same time or in a short time, so that the shopping habit of the customer is expressed.
Will be firstA user is firstThe purchase time of the commodities of the types is arranged in time sequence to obtain the purchase time sequence of each commodity of the typesTo purchase a time seriesTo middleEach elementBy way of example, the element pair acquisition method of (1) is illustrated in time seriesMiddle capture and elementElements of the next positionIn the first placeOne of the users is not the firstAcquisition and element in time series of purchases of individual types of merchandiseWith the smallest spacing between elements as a possible element, all non-firstAll possible elements obtained by the purchase time sequence of the commodities of the types form a possible element set, and it needs to be noted that the purchase time of each element in the possible element set is greater than that of the elementThe time of purchase of (2), the selection of the element in the set of possible elementsAndelements between purchase times and with the smallest purchase timeThe first element of (1), the second elementAnd the first element of the element constitutes an element pair.
The element pair is used for reflecting the purchasing habit of a user for purchasing the commodity, and no other commodity is purchased between the two types of commodities, namely, the commodity of one type is purchased and then the commodity of the other type is purchased by the habit.
Obtain arbitrary twoTwo types of commodities corresponding to each node form a commodity type pair; first, theEach commodity type pair containsType goods andtype of goods according toTime series of purchases of type goods andthe purchase time sequence of the type goods is obtainedA set of pairs of elements for each pair of item types.
According to the firstThe first of the element pair set of each item type pairThe element pairs are each provided with a purchase time intervalTo the secondThe element pairs of each commodity type pair are collected to obtain the first timeA purchase interval histogram for each merchandise type pair, the histogram having an abscissa of a value of a sorted purchase intervalThe ordinate of the histogram is the number value corresponding to each purchase interval time value.
If the number of places where the time between purchases is smaller in the histogram is larger, it indicates that the current customer purchases the two items more frequently in a short time. If the number of places where the purchase interval is larger in the histogram is larger, it means that the current customer has a higher frequency of purchasing the two items with a longer interval.
According to the firstThe histogram of the purchase interval of each commodity type pair is calculated to obtainAverage value of purchase time interval values of individual commodity type pairsBy which value the user is reflected in purchasing the firstThe time interval between the purchase of two commodities in each commodity type pair, i.e. after the user purchases one commodity, the habit is in the intervalThen another corresponding commodity is purchased, and the purchasing habit of the user is reflected by the data to obtain the second commodityThe number of element pairs in the element pair set of each commodity type pair is used as the total numerical value of the purchase time interval value and is recorded as a first quantityThe larger the value is, the more the user is accustomed to purchasingOne of the commodity type pairsAfter the commodity is purchased, another commodity in the commodity type pair is purchased, and a two-dimensional vector formed by the first quantity of the commodity type pair and the average value of the time interval values is recorded as a weight vector:
is shown asFirst completely undirected graph of individual userThe larger the average value of the purchase time interval values of the individual commodity type pairs, the longer the interval time per purchase is. Reflecting a consumption habit.
Is shown asFirst completely undirected graph of individual userThe lower the total number value of the individual item type pair element pair, the lower the possibility of expressing that the current user has a habit of purchasing two items.
By weight vectorAs a firstFirst completely undirected graph of individual userThe edge weight of the connecting line between two commodity types in each commodity type pair.
Step S003: and obtaining the degree of membership among the users according to the user set with various consumption habits of each user.
If the consumption is similar, the relation of partial side weights in the completely undirected graph presents equal ratio or approximate equal ratio to represent the consumption habit approximation, and further the degree of membership of the consumption relation among the users is established, and further the users are classified.
1. And pruning the first completely undirected graph of each user to obtain a second completely undirected graph.
Deleting the edge of the first edge in the first completely undirected graph of each user to obtain a middle completely undirected graph, wherein the first edge is the total numerical value of the purchase time interval values of the edge weights in the undirected graphAnd deleting the isolated nodes in the middle completely undirected graph to obtain a second completely undirected graph, wherein the edges are 0.
2. And obtaining the degree of membership of each user according to the second completely undirected graph.
The number of edges of the second completely undirected graph of each user is obtained, a first user sequence is obtained by arranging all users from small to large according to the number of the edges of the second completely undirected graph of all the users, a set formed by all the users of which the position sequence of each user is greater than that of the user is marked as a user set with various consumption habits of each user, and the user set with various consumption habits of each user is formed by a plurality of users with various consumption habits.
Splitting the second completely undirected graph of the user and the same subgraph in the second completely undirected graph of each diversified consuming habit user to obtain the same subgraph set, wherein the same subgraph refers to a partial graph structure with the same nodes and the same connecting edges of the two second completely undirected graphs, the same connecting edges are the same and do not represent the same edge weight of the connecting edges, and the same subgraph in the same subgraph set forms a subgraph pair to obtain a subgraph setTo a plurality of sub-image pairs, subtracting the edge weight values of the corresponding edges of two sub-images in each sub-image pair to obtain a difference value vector, multiplying the values of two dimensions in the difference value vector to obtain a product value, and taking the product value as the comprehensive difference of each corresponding edge of the sub-image pairAnd carrying out negative correlation mapping processing on the comprehensive difference to obtain a matching value of each corresponding edge of the subgraph, wherein the negative correlation processing calculation formula is as follows:
matching the subgraph to all corresponding edgesAnd calculating the mean value to obtain a first matching value of each sub-graph pair, and taking the mean value of the first matching values of all sub-graph pairs in the same sub-graph set as the degree of membership between the user and each diversified user with consumption habits, namely the degree of membership between users, wherein the greater the degree of membership, the more sub-graphs with similar weights exist between the users, so that the consumption habits between the two users are similar.
Step S004: and constructing a user consumption relation completely undirected graph according to the degree of membership among the users, and obtaining a plurality of user types according to the user consumption relation completely undirected graph.
In a completely undirected graph, if the edge weight value between two nodes is lower, the consumption habit difference corresponding to two users is larger, so that the two nodes are not classified into one type, and if the edge weight value between the two nodes is higher, the consumption habit difference corresponding to two users is smaller, so that the two nodes are classified into one type.
And taking each user as a node, and taking the degree of membership between the users as an edge weight to construct a complete undirected graph, and recording the completely undirected graph as a user consumption relationship. In the completely undirected graph of the user consumption relationship, if the edge weight value between two nodes is lower, the consumption habits of two corresponding users are more similar, and therefore the two users should be classified into one type, and if the edge weight value between two nodes is higher, the consumption habits of two corresponding users are more different, and the two users should not be classified into one type.
And (3) segmenting the completely undirected graph of the user consumption relation by utilizing a graph segmentation method of the maximum stream to obtain a plurality of segmentation blocks, and taking users corresponding to all nodes of each segmentation block as a user type to obtain a plurality of user types.
In summary, according to the embodiments of the present invention, the configuration data is represented according to the historical transaction information of the user, and then the degree of membership of different users in the consumption habits is obtained according to the shopping habit diagram structure data of different users, so that the classification accuracy when classifying the users according to the consumption habits of the users is improved.
It should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. The processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. A method for user classification management based on historical transaction information, the method comprising:
acquiring historical transaction information of each user, wherein the historical transaction information of each user comprises the type of commodities purchased by each user and the purchase time of each type of commodities, and obtaining a first complete undirected graph of each user according to the historical transaction information of each user;
building the first completely undirected graph of each user to obtain a second completely undirected graph of each user, and obtaining a user set with various consumption habits of each user according to the second completely undirected graphs of all the users; obtaining the degree of membership among the users according to the user set with various consumption habits of each user;
and constructing a user consumption relation completely undirected graph according to the degree of membership among the users, and obtaining a plurality of user types according to the user consumption relation completely undirected graph.
2. The method for user classification management based on historical transaction information as claimed in claim 1, wherein the method for obtaining the first completely undirected graph of each user according to the historical transaction information of each user comprises:
acquiring commodity type sets purchased by all users from all user historical transaction information sets, acquiring the purchase time of each user for each type of commodity, randomly combining any two types of commodities in the commodity type sets to obtain a commodity type pair, acquiring an interval time value set of the commodity type pair according to the purchase time of each user for the two types of commodities in each commodity type pair, wherein the interval time value set comprises a plurality of interval time values, counting the plurality of interval time values of the commodity type pair to obtain a purchase interval time histogram, acquiring a time interval value average value of the commodity type pair according to the purchase interval time histogram, and acquiring the total number of the interval time values in the interval time value set as a first number of the commodity type pair; forming a two-dimensional vector by the first quantity of the commodity type pairs and the average value of the time interval values, and recording the two-dimensional vector as a weight vector;
and constructing a first completely undirected graph by taking each type of commodity as a node and taking the weight vector as an edge weight.
3. The method for user category management based on historical transaction information as claimed in claim 1, wherein the method for obtaining the degree of affiliation between users according to each user's set of users with diverse consumption habits comprises:
splitting the second completely undirected graph of the user and the same subgraph in the second completely undirected graph of each various consumption habit user to obtain the same subgraph set, forming a subgraph pair by the same subgraph in the same subgraph set to obtain a plurality of subgraph pairs, subtracting the edge weights of the corresponding edges of the two subgraphs in each subgraph pair to obtain a difference vector, taking a product value obtained by multiplying the values of two dimensions in the difference vector as the comprehensive difference of each corresponding edge of the subgraph pair, obtaining the matching value of each corresponding edge of the subgraph pair according to the comprehensive difference, averaging the matching values of all corresponding edges of the subgraph pair to obtain a first matching value of each subgraph pair, and taking the average value of the first matching values of all subgraph pairs in the same subgraph set as the degree of membership between the user and each various consumption habit user, namely the degree of membership between the users.
4. The method for user category management based on historical transaction information according to claim 1, wherein the method for building the first completely undirected graph of each user to obtain the second completely undirected graph of each user comprises:
deleting the edge of the first edge in the first completely undirected graph of the user to obtain a middle completely undirected graph, wherein the first edge is an edge of which one element of the edge weight in the undirected graph is 0, and deleting the isolated node in the middle completely undirected graph to obtain a second completely undirected graph.
5. The method for user category management based on historical transaction information as claimed in claim 1, wherein the method for obtaining a diverse set of users with consumption habits of each user according to the second completely undirected graph of all users comprises:
the number of edges of the second completely undirected graph of each user is obtained, a first user sequence is obtained by arranging all users from small to large according to the number of the edges of the second completely undirected graph of all the users, a set formed by all the users of which the position sequence of each user is greater than that of the user is marked as a user set with various consumption habits of each user, and the user set with various consumption habits of each user is formed by a plurality of users with various consumption habits.
6. The method for user classification management based on historical transaction information according to claim 1, wherein the formula for constructing the completely undirected graph of the user consumption relationship according to the degree of membership between users comprises:
and taking each user as a node, and taking the degree of membership between the users as an edge weight to construct a complete undirected graph, and recording the completely undirected graph as a user consumption relationship.
7. The method for user category management based on historical transaction information as claimed in claim 1, wherein the method for obtaining multiple user types according to the completely undirected graph of user consumption relationship comprises:
and segmenting the user consumption relation completely undirected graph by utilizing the maximum stream to obtain a plurality of segmentation blocks, and taking users corresponding to all nodes of each segmentation block as a user type to obtain a plurality of user types.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211330543.0A CN115392406B (en) | 2022-10-28 | 2022-10-28 | User classification management method based on historical transaction information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211330543.0A CN115392406B (en) | 2022-10-28 | 2022-10-28 | User classification management method based on historical transaction information |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115392406A true CN115392406A (en) | 2022-11-25 |
CN115392406B CN115392406B (en) | 2023-04-28 |
Family
ID=84115020
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211330543.0A Active CN115392406B (en) | 2022-10-28 | 2022-10-28 | User classification management method based on historical transaction information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115392406B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003099643A (en) * | 2001-09-21 | 2003-04-04 | Ricoh Co Ltd | On-line shopping system and on-line shopping method |
KR100824197B1 (en) * | 2007-05-04 | 2008-04-21 | 정해봉 | System and method for offering multimedia shopping contents |
CN103544632A (en) * | 2013-07-22 | 2014-01-29 | 杭州师范大学 | Method and system for individually recommending network commodities |
CN105427125A (en) * | 2015-10-29 | 2016-03-23 | 电子科技大学 | Goods clustering method based on goods network connection graph |
CN105931066A (en) * | 2015-09-24 | 2016-09-07 | 中国银联股份有限公司 | Transaction data processing method and device |
CN106529968A (en) * | 2016-09-29 | 2017-03-22 | 深圳大学 | Customer classification method and system thereof based on transaction data |
CN109191240A (en) * | 2018-08-14 | 2019-01-11 | 北京九狐时代智能科技有限公司 | A kind of method and apparatus carrying out commercial product recommending |
CN110503497A (en) * | 2018-05-16 | 2019-11-26 | 江苏天智互联科技股份有限公司 | A kind of electric business management platform Method of Commodity Recommendation of the consumption habit based on client |
CN112131471A (en) * | 2020-09-23 | 2020-12-25 | 平安国际智慧城市科技股份有限公司 | Method, device, equipment and medium for relation recommendation based on unauthorized undirected graph |
CN112380453A (en) * | 2021-01-15 | 2021-02-19 | 腾讯科技(深圳)有限公司 | Article recommendation method and device, storage medium and equipment |
-
2022
- 2022-10-28 CN CN202211330543.0A patent/CN115392406B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003099643A (en) * | 2001-09-21 | 2003-04-04 | Ricoh Co Ltd | On-line shopping system and on-line shopping method |
KR100824197B1 (en) * | 2007-05-04 | 2008-04-21 | 정해봉 | System and method for offering multimedia shopping contents |
CN103544632A (en) * | 2013-07-22 | 2014-01-29 | 杭州师范大学 | Method and system for individually recommending network commodities |
CN105931066A (en) * | 2015-09-24 | 2016-09-07 | 中国银联股份有限公司 | Transaction data processing method and device |
CN105427125A (en) * | 2015-10-29 | 2016-03-23 | 电子科技大学 | Goods clustering method based on goods network connection graph |
CN106529968A (en) * | 2016-09-29 | 2017-03-22 | 深圳大学 | Customer classification method and system thereof based on transaction data |
CN110503497A (en) * | 2018-05-16 | 2019-11-26 | 江苏天智互联科技股份有限公司 | A kind of electric business management platform Method of Commodity Recommendation of the consumption habit based on client |
CN109191240A (en) * | 2018-08-14 | 2019-01-11 | 北京九狐时代智能科技有限公司 | A kind of method and apparatus carrying out commercial product recommending |
CN112131471A (en) * | 2020-09-23 | 2020-12-25 | 平安国际智慧城市科技股份有限公司 | Method, device, equipment and medium for relation recommendation based on unauthorized undirected graph |
CN112380453A (en) * | 2021-01-15 | 2021-02-19 | 腾讯科技(深圳)有限公司 | Article recommendation method and device, storage medium and equipment |
Non-Patent Citations (2)
Title |
---|
NATALIA ET AL: "Segmentation of Mobile Applications Users Classification of Customer E-Loyalty Behavior in Online Shopping Platform", 《2020 INTERNATIONAL CONFERENCE ON INFORMATION MANAGEMENT AND TECHNOLOGY (ICIMTECH)》 * |
郑晓薇等: "基于FCM算法的电子商务客户分类研究", 《微型机与应用》 * |
Also Published As
Publication number | Publication date |
---|---|
CN115392406B (en) | 2023-04-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
TWI587229B (en) | Push method and device for product information | |
CN110032606B (en) | Sample clustering method and device | |
Romdhane et al. | An efficient approach for building customer profiles from business data | |
CN102609854A (en) | Client partitioning method and device based on unified similarity calculation | |
CN106127493A (en) | A kind of method and device analyzing customer transaction behavior | |
CN108053268A (en) | A kind of commercial articles clustering confirmation method and device | |
CN106250431A (en) | A kind of Color Feature Extraction Method based on classification clothing and costume retrieval system | |
WO2017119952A1 (en) | Consumer decision tree generation system | |
CN108694606A (en) | A kind of customer portrait and service push method based on correlation rule | |
CN105956122A (en) | Object attribute determining method and device | |
CN115797005A (en) | Prediction model training method, prediction model training device, prediction model commodity recommendation method, prediction model training device, commodity recommendation device, electronic equipment and medium | |
US20160171365A1 (en) | Consumer preferences forecasting and trends finding | |
CN112560105B (en) | Joint modeling method and device for protecting multi-party data privacy | |
Akgüller et al. | Geodetic convex boundary curvatures of the communities in stock market networks | |
Azadnia et al. | Integration model of Fuzzy C means clustering algorithm and TOPSIS Method for Customer Lifetime Value Assessment | |
CN115392406A (en) | User classification management method based on historical transaction information | |
CN111340566A (en) | Commodity classification method and device, electronic equipment and storage medium | |
CN111651456B (en) | Potential user determination method, service pushing method and device | |
CN110969491B (en) | Commodity pushing method, system and equipment based on network path | |
CN110427558B (en) | Resource processing event pushing method and device | |
CN111784403A (en) | User category analysis method and device based on online shopping mall and computer equipment | |
CN114519600A (en) | Graph neural network CTR estimation algorithm fusing adjacent node variances | |
Mehrazma et al. | Customer Clustering by Combining the Particle Swarm And K-Means Algorithms and Analyzing Their Behavior on Commercial Websites | |
Huang et al. | Clustering analysis on e-commerce transaction based on k-means clustering | |
CN110751180A (en) | False comment group division method based on spectral clustering |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |