CN115983900A - Method, apparatus, device, medium, and program product for constructing user marketing strategy - Google Patents
Method, apparatus, device, medium, and program product for constructing user marketing strategy Download PDFInfo
- Publication number
- CN115983900A CN115983900A CN202310042183.2A CN202310042183A CN115983900A CN 115983900 A CN115983900 A CN 115983900A CN 202310042183 A CN202310042183 A CN 202310042183A CN 115983900 A CN115983900 A CN 115983900A
- Authority
- CN
- China
- Prior art keywords
- user
- index
- transaction
- value
- subdivision
- 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
- 238000000034 method Methods 0.000 title claims abstract description 89
- 238000010276 construction Methods 0.000 claims abstract description 33
- 238000003860 storage Methods 0.000 claims abstract description 19
- 238000013486 operation strategy Methods 0.000 claims abstract description 10
- 239000011159 matrix material Substances 0.000 claims description 57
- 230000000694 effects Effects 0.000 claims description 50
- 238000010606 normalization Methods 0.000 claims description 27
- 238000004590 computer program Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 23
- 238000004364 calculation method Methods 0.000 claims description 20
- 238000007781 pre-processing Methods 0.000 claims description 14
- 238000012216 screening Methods 0.000 claims description 14
- 239000002131 composite material Substances 0.000 claims description 13
- 238000007635 classification algorithm Methods 0.000 claims description 9
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000000638 solvent extraction Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 2
- 230000005484 gravity Effects 0.000 claims description 2
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 230000000007 visual effect Effects 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 17
- 238000011156 evaluation Methods 0.000 description 11
- 230000015654 memory Effects 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 238000004891 communication Methods 0.000 description 8
- 238000007477 logistic regression Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 230000011218 segmentation Effects 0.000 description 4
- 238000012549 training Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000013475 authorization Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000004806 packaging method and process Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000004936 stimulating effect Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The disclosure provides a construction method of a user marketing strategy, which can be applied to the technical field of artificial intelligence. The method comprises the following steps: the construction method of the user marketing strategy comprises the following steps: acquiring user transaction data; constructing a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing the transaction condition and the transaction capability of a user within a preset time range; establishing an improved user clustering model based on the user portrait index system, wherein the improved user clustering model is established based on an RFM (remote visual model), and comprises a plurality of comprehensive indexes, wherein each comprehensive index comprises the same or different numbers of subdivision indexes; dividing the user group based on the improved user group model; and constructing a user classification operation strategy based on the user group division result. The disclosure also provides a construction device, equipment, a storage medium and a program product of the user marketing strategy.
Description
Technical Field
The present disclosure relates to the field of artificial intelligence technologies, and in particular, to a method, an apparatus, a device, a medium, and a program product for constructing a user marketing strategy.
Background
The construction of user marketing strategies is crucial to business development. At present, user loss prediction and marketing are often performed based on user images. The method comprises the following basic steps: acquiring various kinds of relevant information of a user; mapping the user information to a user portrait to obtain user characteristics, and performing system description on the user information in a specific service scene; and after the user characteristics are obtained, predicting the user loss condition by using a machine learning model, and marketing the user with high user loss probability. However, in the above method, due to the individual differences of the users, the user figures described in a unified manner are often insufficient to represent the characteristics of the users, so that the marketing strategy is not targeted enough, and the collected user information data is not fully utilized, which results in the waste of computing resources.
Disclosure of Invention
In view of the foregoing, embodiments of the present disclosure provide a method, an apparatus, a device, a medium, and a program product for constructing a user marketing strategy, which improve the pertinence of the user marketing strategy and fully utilize data resources.
According to a first aspect of the present disclosure, there is provided a method for constructing a user marketing strategy, including: acquiring user transaction data; constructing a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing the transaction condition and the transaction capability of a user within a preset time range; establishing an improved user clustering model based on the user portrait index system, wherein the improved user clustering model is established based on an RFM (remote object model), and comprises a plurality of comprehensive indexes, wherein each comprehensive index comprises the same or different numbers of subdivision indexes; dividing the user group based on the improved user group model; and constructing a user classification operation strategy based on the user group division result.
According to an embodiment of the disclosure, the user representation index system includes a plurality of user representation dimensions, wherein each user representation dimension includes a plurality of user transaction data types, the composite index is built based on a combination of the user transaction data types, and the segment index is associated with the user transaction data types.
According to an embodiment of the disclosure, the user portrait dimension includes user transaction habits, user transaction preferences and user value, wherein the user transaction habits include user transaction frequency data and user transaction willingness data; and/or the user transaction preference comprises the interest degree of a user product and the acceptance degree of a user marketing activity; and/or the user value includes user asset data and user transaction amount data.
According to the embodiment of the disclosure, the user transaction frequency data comprises the number of continuous transaction days of the user and the number of transaction times of the user in one year, and the user transaction intention data comprises proportion data of the monthly transaction amount of the user and the monthly income of the user; and/or, the user asset data comprises user annual income; and/or the user transaction amount data comprises a user annual transaction amount and a user historical average transaction amount.
According to an embodiment of the disclosure, the composite index includes a user purchasing power score, a user activity score, and a user churn probability.
According to the embodiment of the disclosure, the subdivision indexes in the user purchasing power score comprise user annual transaction amount, user total transaction amount and user annual income; and/or the segment indexes in the activity degree score comprise the annual transaction times of the user, the average transaction interval days of the user and the continuous transaction days of the user; and/or the segment indexes in the attrition probability comprise the number of continuous transaction days of the user, the number of transactions within one year of the user, the interest degree of the user product and the acceptance degree of the user marketing activity.
According to the embodiment of the disclosure, the user purchasing power score and the user activity score are calculated based on a Topsis method and an entropy weight method; and/or the user churn probability is calculated based on a classification algorithm.
According to the embodiment of the disclosure, the user purchasing power score and the user activity score are calculated based on a Topsis method and an entropy weight method, and the calculation comprises the following steps: acquiring user subdivision index values, and constructing a forward matrix based on the user subdivision index values, wherein the user subdivision index values comprise m subdivision index values corresponding to n users, and the forward matrix corresponds to user purchasing power scores or user activity scores; carrying out standardization processing on the forward matrix to obtain a standardized matrix; calculating the information entropy and the information utility value of each subdivision index based on the standardized matrix, and acquiring the entropy weight of each subdivision index; and calculating the purchasing power scoring value and/or the activity scoring value of the user based on the standardized matrix and the entropy weight of each subdivision index.
According to an embodiment of the present disclosure, calculating a user purchasing power rating value and/or a user activity rating value based on the normalization matrix and the entropy weight of each segment index includes: respectively calculating the jth subdivision index maximum value and the jth subdivision index minimum value based on the standardized matrix; calculating the maximum scoring distance of the ith user based on the normalization matrix, the entropy weight of each subdivision index and the maximum value of each subdivision index; calculating the minimum scoring distance of the ith user based on the normalization matrix, the entropy weight of each subdivision index and the minimum value of each subdivision index; and calculating a user score value of the ith user based on the maximum scoring distance of the ith user and the minimum scoring distance of the ith user, wherein the user score value is a user purchasing power score value or a user activity score value.
According to an embodiment of the present disclosure, the classification algorithm comprises a logistic regression algorithm.
According to an embodiment of the present disclosure, the dividing the user group based on the improved user grouping model includes: constructing a user grouping coordinate system based on the comprehensive indexes, wherein the dimensionality of the user grouping coordinate system is the same as the number of the comprehensive indexes; acquiring the average index value of each comprehensive index in the improved user clustering model based on user transaction data; and quadrant division is carried out on the user grouping coordinate system based on the average index value of each comprehensive index, and user subdivision groups are obtained.
According to an embodiment of the present disclosure, the constructing a user portrait index system based on the user transaction data includes: pre-processing the user transaction data, the pre-processing comprising: screening, cleaning, structuring, counting and abnormal value processing; and classifying and screening the preprocessed user transaction data to construct the user portrait index system.
A second aspect of the present disclosure provides a device for constructing a user marketing strategy, including: an acquisition module configured to acquire user transaction data; the first construction module is configured to construct a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing the transaction condition and the transaction capability of a user within a preset time range; the second construction module is configured to establish an improved user clustering model based on the user portrait index system, the improved user clustering model is constructed based on an RFM model, and the improved user clustering model comprises a plurality of comprehensive indexes, wherein each comprehensive index comprises the same or different numbers of subdivision indexes; a dividing module configured to divide a user group based on the improved user grouping model; and the strategy construction module is configured to construct a user classification operation strategy based on the user group division result.
According to an embodiment of the present disclosure, the first building module includes a preprocessing submodule and a screening submodule. Wherein a pre-processing submodule is configured to pre-process the user transaction data, the pre-processing comprising: screening, cleaning, structuring, statistics, and outlier processing. And the screening submodule is configured to classify and screen the preprocessed user transaction data to construct the user portrait index system.
According to an embodiment of the present disclosure, the policy building module includes a coordinate system establishing submodule, a first calculating submodule, and a second calculating submodule. The coordinate system establishing submodule is configured to establish a user clustering coordinate system based on the comprehensive indexes, and the dimensionality of the user clustering coordinate system is the same as the number of the comprehensive indexes. The first calculation submodule is configured to obtain an average index value of each comprehensive index in the improved user clustering model based on user transaction data. And the second calculation submodule is configured to perform quadrant division on the user grouping coordinate system based on the average index value of each comprehensive index to obtain a user subdivision group.
According to an embodiment of the disclosure, the second construction module includes a forward quantization submodule, a normalization submodule, an entropy weight calculation submodule, and a score value calculation submodule. The forward sub-module is configured to obtain user segment index values, and construct a forward matrix based on the user segment index values, wherein the user segment index values comprise m segment index values corresponding to n users, and the forward matrix corresponds to a user purchasing power score or a user activity score. And the normalization submodule is configured to normalize the forward matrix to obtain a normalized matrix. The entropy weight calculation sub-module is configured to calculate the information entropy and the information utility value of each subdivision index based on the standardized matrix, and obtain the entropy weight of each subdivision index. And the credit value calculation submodule is configured to calculate a user purchasing power credit value and/or a user activity credit value based on the standardization matrix and the entropy weight of each subdivision index.
A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of constructing a user marketing strategy described above.
The fourth aspect of the present disclosure also provides a computer-readable storage medium having stored thereon executable instructions, which when executed by a processor, cause the processor to execute the above construction method of the user marketing strategy.
The fifth aspect of the present disclosure also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the computer program implements the method for constructing the user marketing strategy.
According to the method provided by the embodiment of the disclosure, after the user portrait is constructed by using the user transaction data, the improved user clustering model is further constructed by using the user portrait, and the user population is divided based on the improved user clustering model, so that the user characteristic data can be fully utilized to segment the user population, the user marketing strategy is constructed based on the pertinence of different user segmentation populations, the data utilization rate is improved, and the user marketing effect is favorably optimized.
Drawings
The foregoing and other objects, features and advantages of the disclosure will be apparent from the following description of embodiments of the disclosure, which proceeds with reference to the accompanying drawings, in which:
fig. 1 schematically shows an application scenario diagram of a construction method, an apparatus, a device, a medium, and a program product of a user marketing strategy according to an embodiment of the present disclosure.
Fig. 2 schematically shows a flowchart of a method of constructing a user marketing strategy according to an embodiment of the present disclosure.
FIG. 3 schematically illustrates a flow chart of a method of building a user portrait index hierarchy based on the user transaction data, in accordance with an embodiment of the present disclosure.
Fig. 4 schematically illustrates a flowchart of a method for calculating a user purchasing power score or a user activity score based on a Topsis method combined with an entropy weight method according to an embodiment of the present disclosure.
Fig. 5 is a flowchart schematically illustrating a method for calculating a user purchasing power rating value and/or a user activity rating value based on the normalization matrix and the entropy weight of each segment index according to an embodiment of the present disclosure.
FIG. 6 schematically shows a flow chart of a method for user population partitioning based on the improved user clustering model according to an embodiment of the present disclosure.
Fig. 7 is a block diagram schematically illustrating a construction apparatus of a user marketing strategy according to an embodiment of the present disclosure.
Fig. 8 schematically shows a block diagram of a first building block according to an embodiment of the present disclosure.
FIG. 9 schematically shows a block diagram of a policy building module according to an embodiment of the present disclosure.
Fig. 10 schematically shows a block diagram of a second building block according to an embodiment of the present disclosure.
FIG. 11 schematically illustrates a block diagram of an electronic device suitable for implementing a construction method of a user marketing strategy according to an embodiment of the present disclosure.
Detailed Description
Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood that these descriptions are illustrative only and are not intended to limit the scope of the present disclosure. In the following detailed description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the disclosure. It may be evident, however, that one or more embodiments may be practiced without these specific details. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present disclosure.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "A, B and at least one of C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include, but not be limited to, systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).
In business development, it is important to make a suitable user marketing strategy. In the prior art, a method with higher popularity is to predict and market user loss based on user images. The method comprises the following basic steps: acquiring user related information; mapping the user information to a user portrait to obtain user characteristics, and performing system description on the user information in a specific service scene; after the user characteristics are obtained, predicting the user loss condition by using a machine learning model, and marketing and recovering the users with high user loss probability. However, in the above method, due to the individual differences of the users, the user figures described uniformly are often not enough to represent the characteristics of the users, so that the marketing strategy is not targeted. On the other hand, since the collected user-related information is not further integrated, the above method cannot fully utilize the collected user information data, resulting in waste of data resources.
In view of the foregoing problems in the prior art, an embodiment of the present disclosure provides a method for constructing a user marketing strategy, including: acquiring user transaction data; constructing a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing the transaction condition and the transaction capability of a user within a preset time range; establishing an improved user clustering model based on the user portrait index system, wherein the improved user clustering model is established based on an RFM (remote visual model), and comprises a plurality of comprehensive indexes, wherein each comprehensive index comprises the same or different numbers of subdivision indexes; dividing the user group based on the improved user group model; and constructing a user classification operation strategy based on the user group division result.
According to the method provided by the embodiment of the disclosure, after the user portrait is constructed by using the user transaction data, the improved user group model is further constructed by using the user portrait, and the user group division is performed based on the improved user group model, so that the user characteristic data can be fully utilized to subdivide the user group, the user marketing strategy is constructed based on the pertinence of different user subdivision groups, the data utilization rate can be greatly improved, and the optimization of the user marketing effect is facilitated.
In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure, application and other processing of the personal information of the related user are all in accordance with the regulations of related laws and regulations, necessary confidentiality measures are taken, and the customs of the public order is not violated.
In the technical scheme of the disclosure, before the personal information of the user is obtained or collected, the authorization or the consent of the user is obtained.
It should be noted that the construction method, apparatus, device, medium, and program product of the user marketing strategy provided in the embodiments of the present disclosure may be applied to the artificial intelligence technology in the aspects related to the classification of user customers and marketing, and may also be applied to various fields other than the artificial intelligence technology, such as the financial field. The application fields of the method, the device, the equipment, the medium and the program product for constructing the user marketing strategy provided by the embodiment of the disclosure are not limited.
The above-described operations for carrying out at least one of the objects of the present disclosure will be described with reference to the accompanying drawings and description thereof.
Fig. 1 schematically shows an application scenario diagram of a construction method, an apparatus, a device, a medium, and a program product of a user marketing strategy according to an embodiment of the present disclosure.
As shown in fig. 1, the application scenario 100 according to this embodiment may include terminal devices 1, 102, 103, a network 104, and a server 105. The network 104 serves as a medium for providing communication links between the terminal devices 101, 102, 103 and the server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. The terminal devices 101, 102, 103 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).
The terminal devices 101, 102, 103 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 105 may be a server providing various services, such as a background management server (for example only) providing support for websites browsed by users using the terminal devices 101, 102, 103. The background management server may analyze and perform other processing on the received data such as the user request, and feed back a processing result (e.g., a webpage, information, or data obtained or generated according to the user request) to the terminal device.
It should be noted that the method for constructing the user marketing strategy provided by the embodiment of the present disclosure may be generally executed by the server 105. Accordingly, the construction device of the user marketing strategy provided by the embodiment of the present disclosure can be generally disposed in the server 105. The method for constructing the user marketing strategy provided by the embodiment of the present disclosure may also be executed by a server or a server cluster which is different from the server 105 and can communicate with the terminal devices 101, 102, 103 and/or the server 105. Accordingly, the constructing apparatus of the user marketing strategy provided by the embodiment of the present disclosure may also be disposed in a server or a server cluster different from the server 105 and capable of communicating with the terminal devices 101, 102, 103 and/or the server 105.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
The following describes in detail a method for constructing a user marketing strategy according to the disclosed embodiment with reference to fig. 2 to 6 based on the scenario described in fig. 1.
Fig. 2 schematically shows a flowchart of a method of constructing a user marketing strategy according to an embodiment of the present disclosure.
As shown in fig. 2, the method for constructing the user marketing strategy according to this embodiment includes operations S210 to S250, and the method for constructing the user marketing strategy may be executed by a processor, or may be executed by any electronic device including a processor.
In operation S210, user transaction data is acquired.
According to embodiments of the present disclosure, user transaction data includes raw transaction data that may be collected that may be used to construct a user representation. It should be appreciated that user transaction data may be associated with a particular business scenario to accommodate the construction of user marketing strategies under different business scenarios. As a typical example, in a credit card marketing service, user transaction data may be data consumed using a credit card. It will be appreciated that the credit card consumption data may include time of credit card consumption, merchant, amount, etc.
In operation S220, a user portrait index system is constructed based on the user transaction data, and the user portrait index system is used to represent a transaction situation and a transaction capability of a user within a preset time range.
In an embodiment of the present disclosure, the user transaction data includes user transaction raw data. The user portrait index system can be constructed by processing the user transaction raw data. For example, the user representation can be constructed by processing the credit card consumption data to obtain data such as user consumption frequency, consumption amount, consumption preference, consumption habit and the like.
FIG. 3 schematically shows a flow diagram of a method of building a user portrait index system based on the user transaction data, in accordance with an embodiment of the present disclosure.
As shown in FIG. 3, the method for constructing a user portrait index system based on the user transaction data of this embodiment includes operations S310 to S320.
In operation S310, the user transaction data is preprocessed. Wherein the pre-processing comprises: screening, cleaning, structuring, statistics, and outlier processing.
In operation S320, the preprocessed user transaction data is classified and filtered, and the user portrait index system is constructed. It can be understood that the user transaction data can be further classified, screened and extracted to obtain classified user transaction information after being preprocessed, so that the user portrait can be constructed.
In some embodiments, the user representation index system includes a plurality of user representation dimensions, wherein each user representation dimension includes a plurality of user transaction data types, the composite index is built based on the user transaction data types in combination, and the segment index is associated with the user transaction data types. It is understood that the segment index may be a user transaction data type, or may be constructed by processing calculation based on the user transaction data type. The comprehensive index is a further combination of the subdivision indexes based on preset rules, and is used for fully reflecting the value of the user subdivision groups.
In some embodiments, the user representation dimensions include user transaction habits, user transaction preferences, and user value. The user transaction habits comprise user transaction frequency data and user transaction intention data; and/or the user transaction preference comprises user product interest degree and user marketing activity acceptance degree; and/or the user value includes user asset data and user transaction amount data.
In some preferred embodiments, the user transaction frequency data includes the number of continuous transaction days of the user and the number of transactions within one year of the user, and the user transaction intention data includes specific gravity data of the user monthly transaction amount and the user monthly income; and/or, the user asset data comprises user annual income; and/or the user transaction amount data comprises a user annual transaction amount and a user historical average transaction amount.
For example, by extracting the consumption time of the credit card consumption data of the user a within one year, the consumption frequency, the continuous consumption duration, the consumption times and other data can be analyzed, so as to further obtain the portrait dimension data of the user consumption habit, the consumption preference and the like. Similarly, the consumption ability and the consumption habit of the user can be known through the consumption amount of the user. Furthermore, the consumption habit, the consumption capability, the consumption preference and the like of the portrait dimension data are obtained.
In one specific example, a credit card consumption scenario is taken as an example, and a three-dimensional user representation index system can be constructed through credit card consumption data. Including consumption habits, consumption preferences and user value. The consumption habit includes user transaction data types including continuous consumption days, consumption times in one year and monthly income proportion of monthly consumption. Specifically, if the interval between the two previous and subsequent consumption card swiping records is less than 30 days, the user is considered to continuously consume in the time period. The consumption preferences include user transaction data types including a user's level of interest in a financial product and a user's willingness to accept a credit card marketing campaign. Wherein, the user's interest level in the financial product takes the total number of times of history of purchasing the financial product as the interest level value. Specific evaluation criteria for the willingness of the user to accept the credit card marketing campaign can be set in a value assignment way. For example, if the user participates in a credit card marketing campaign, the value is set to 3; if the financial product is purchased but not participating in the credit card marketing campaign, setting the value to be 2; if the financial product is not purchased yet after participating in the credit card marketing campaign, the value is set to 1. The user value comprises the annual income of the user, the annual total consumption amount of the user and the average consumption amount of the user in all years. Wherein, the annual income of the user can select the middle value of the income interval filled by the user as the index value; the total annual consumption amount of the user can be the total annual consumption amount of the user, which is 12 months before the current calculation date; the average consumption amount of the user over the years can be obtained and calculated based on the preset time range.
In operation S230, an improved user clustering model is built based on the user portrait index system, the improved user clustering model being built based on an RFM model, the improved user clustering model including a plurality of synthetic indexes, wherein each synthetic index includes the same or different number of segment indexes.
In some embodiments, the composite indicators include a user buying power score, a user activity score, and a user churn probability.
In some embodiments, the segment indicators in the user purchasing power score include a user annual transaction amount, a user total transaction amount, and a user annual revenue; and/or the segment indexes in the activity degree score comprise the annual transaction times of the user, the average transaction interval days of the user and the continuous transaction days of the user; and/or the segment indexes in the attrition probability comprise the number of continuous transaction days of the user, the number of transactions within one year of the user, the interest degree of the user product and the acceptance degree of the user marketing activity. In the above-described example of credit card consumption, the selected segment indicators may be the user annual consumption amount, the user total consumption amount and the user annual income for the user purchasing power score comprehensive indicator. For the user activity rating, the selected subdivision indexes can be the annual consumption times of the user, the average consumption interval days of the user and the continuous consumption days of the user. For the user loss probability, the selected segmentation indexes can be the number of continuous consumption days of the user, the consumption times of the user in one year, the interest degree of the user in financial products and the willingness of the user to accept credit card marketing activities.
In a preferred implementation of the present disclosure, an improved user clustering model is obtained through further integration of user transaction data types. The original three indexes are as follows: r consumes F consumption frequency M consumption amount for the last time, and correspondingly changes R user loss probability F user activity M user purchasing power, so that the data utilization rate is improved, and the model is more comprehensive on customer grouping.
In operation S240, user group division is performed based on the improved user group division model.
According to the embodiment of the disclosure, an improved user clustering model can be further established through a user portrait index system. The classic user clustering model, namely the RFM model, clusters users based on 3 indexes of Recency (Recency), frequency (Frequency) and amount (money) to find out users with potential values, thereby assisting business decisions and improving marketing efficiency. Wherein, the recency represents the last consumption, the frequency represents the consumption frequency, and the amount represents the consumption amount. The three indexes can reflect the user value to a certain extent, but the utilization data is single and difficult to reflect the subdivision characteristics of the user group. In the embodiment of the disclosure, by constructing a user portrait index system, user portrait data which can reflect user transaction characteristics better can be extracted, and the user portrait data is further used as a segmentation index to construct a comprehensive index so as to replace the proximity, frequency and limit indexes in the traditional RFM model, thereby realizing full utilization of data and improving the accuracy of user grouping.
In operation S250, a user classification operation policy is constructed based on the user group division result.
After the user groups are further subdivided based on the improved user grouping model, classified operation strategies can be established for different subdivided groups so as to realize accurate marketing. For example, the users may be classified into high-value, general-value, and low-value users based on the group segmentation results of the users, and may also be classified into to-be-developed, to-be-maintained, to-be-restored clients, and the like. Aiming at different users, different operation strategies are adopted to meet the service requirements and improve the user experience.
It should be noted that, in the embodiment of the present disclosure, before obtaining the information of the user, the consent or authorization of the user may be obtained. For example, a request to obtain user transaction data may be issued to the user prior to operation S210. In case the user agrees or authorizes that the user transaction data can be acquired, the operation S210 is performed.
According to a preferred embodiment of the present disclosure, in the selected composite indicator, the user purchasing power score and the user activity score may be calculated based on a Topsis method combined with an entropy weight method. In a preferred embodiment of the present disclosure, for the user purchasing power score comprehensive index and the user activity score index, the selected segment indexes include a typical maximum index and a minimum index. For example, the annual transaction amount of the user, the total transaction amount of the user, the annual income of the user, the annual transaction times of the user and the continuous transaction days of the user are typical maximum indexes, i.e. the larger the numerical value is, the better the numerical value is; the average number of days between transactions for a user is typically a very small indicator, i.e., the smaller the number the better. In order to form a scientific evaluation quantification system, in a preferred embodiment of the disclosure, a Topsis method is combined with an entropy weight method to calculate a user purchasing power score and a user activity score. The Topsis method is an effective method in multi-objective decision analysis, and is also called as a good-bad solution distance method. The basic principle is that the evaluation objects are sequenced by detecting the distances between the evaluation objects and the optimal solution and the worst solution, and if the evaluation objects are closest to the optimal solution and are also furthest away from the worst solution, the evaluation objects are the best; otherwise it is not optimal. Wherein each index value of the optimal solution reaches the optimal value of each evaluation index. And all the index values of the worst solution reach the worst value of all the evaluation indexes. The Topsis method is used for evaluating the comprehensive index of the user purchasing power score and the comprehensive index of the user activity score, and the extremely large and small indexes contained in the Topsis method and the comprehensive index of the user activity score can be effectively utilized. Further, because different subdivision indexes have different influence degrees on the comprehensive index, the embodiment of the disclosure introduces an entropy weight method to improve the Topsis method, and introduces an entropy weight to each index. The entropy weight method determines the objective weight according to the index variability. If the information entropy of a certain index is smaller, the index is worth of being varied to a greater extent, the more information is provided, the greater the effect can be played in comprehensive evaluation, and the greater the weight is. Conversely, the larger the information entropy of a certain index is, the smaller the degree of variation of the index value is, the smaller the amount of information provided is, the smaller the role played in the comprehensive evaluation is, and the smaller the weight thereof is. Therefore, the smaller the variation degree of the same index in the sample is, the less the information amount reflected by the index is, and the lower the corresponding entropy weight is.
According to the embodiment of the disclosure, in the selected comprehensive index, the user loss probability is calculated based on a classification algorithm. Typical classification algorithms may include K-nearest neighbor algorithms, decision tree algorithms, naive bayes algorithms, logistic regression algorithms, support vector machine algorithms, random forest algorithms, and the like. The prediction result can be quickly obtained by calculating the user loss probability by using a classification algorithm, and the data processing efficiency is improved.
In some embodiments, the classification algorithm comprises a logistic regression algorithm (LR). The logistic regression algorithm is used as a classical binary classification algorithm in the classification algorithm, has a high calculation speed, can quickly utilize input data, and obtains a prediction result under the condition of occupying small calculation resources.
In the exemplary scenario of credit card marketing as above, the past year credit card customer data is chosen as the training set training model. The number of the training set samples reaches more than 20% of the sample set. In machine learning, the linear regression model is written as:
where y is the prediction function, w is the model parameter, and x is the feature input.
The LR equation construction is based on the second assumption that the probability of a sample is Sigmoid function:
and combining the linear regression model with the Sigmond function to obtain a logistic regression equation, wherein the probability formula of the LR sample is as follows:
where x is the feature input and w is a parameter, also called weight vector, w T x is the inner product of w and x.
Write together:
P(y|x;w)=σ(x) y (1-σ(x)) 1-y
if the probability category threshold is 0.5, then:
if P (y | x; w) is more than or equal to 0.5, then the P belongs to the positive class (y = 1);
if P (y | x; w) < 0.5, it belongs to the negative class (y = 0).
The iterative formula for training LR by adopting the gradient descent method is as follows:
wherein y is (i) A label value, x, representing the ith sample (i) Representing the feature vector of the ith sample. Alpha is a learning rate which is self-set,for the difference of the sample value and the pre-evaluation value>Is the jth feature of the ith sample.
Therefore, after the characteristics of the continuous consumption days, the consumption times in one year, the interest degree of financial products and the willingness of credit card marketing activities are input, the loss probability and the loss probability of the user in one year can be finally output.
Fig. 4 schematically shows a flowchart of a method for calculating a user purchasing power score or a user activity score based on the Topsis method in combination with the entropy weight method according to an embodiment of the present disclosure.
As shown in fig. 4, the method for calculating the user purchasing power score or the user activity score based on the Topsis method in combination with the entropy weight method in this embodiment includes operations S410 to S440.
In operation S410, user segment index values are obtained, and a forward matrix is constructed based on the user segment index values, wherein the user segment index values include m segment index values corresponding to n users, and the forward matrix corresponds to a user purchasing power score or a user activity score.
In operation S420, the normalization matrix is normalized to obtain a normalized matrix.
In operation S430, the information entropy and the information utility value of each subdivision index are calculated based on the normalization matrix, and an entropy weight of each subdivision index is obtained.
In operation S440, a user purchasing power rating value and/or a user activity rating value is calculated based on the normalization matrix and the entropy weight of each segment index.
According to the embodiment of the disclosure, it should be understood that for the user purchasing power score and the user activity score, a forward matrix may be constructed based on specifically selected user segment index values, respectively, so as to realize calculation of the score value.
FIG. 5 is a flow chart that schematically illustrates a method for calculating a user purchasing power rating value and/or a user activity rating value based on the normalization matrix and the entropy weight of each segment indicator, in accordance with an embodiment of the present disclosure.
As shown in fig. 5, the method for calculating the user purchasing power rating value and/or the user activity rating value based on the normalization matrix and the entropy weight of each segment index of the embodiment includes operations S510 to S540.
In operation S510, a jth subdivision index maximum value and a jth subdivision index minimum value are respectively calculated based on the normalization matrix.
In operation S520, the maximum scoring distance of the ith user is calculated based on the normalization matrix, the entropy weight of each refinement index, and the maximum value of each refinement index.
In operation S530, a minimum scoring distance of the ith user is calculated based on the normalization matrix, the entropy weight of each refinement index, and a minimum value of each refinement index.
In operation S540, a user rating value of the ith user is calculated based on the maximum scoring distance of the ith user and the minimum scoring distance of the ith user, wherein the user rating value is a user purchasing power rating value or a user activity rating value.
In some specific embodiments of the present disclosure, the specific process of calculating the user purchasing power score and the user activity score based on the Topsis method in combination with the entropy weight method includes steps S1 to S4.
And S1, carrying out forward processing on the original matrix, and unifying all indexes into a maximum index, wherein the original matrix is the original matrix constructed by the user subdivision index value.
The formula for the conversion from the very small index to the very large index in step S1 is:
X i ′=max{X 1 ,X 2 ,...X i }-X i 。
and S2, constructing a forward matrix, and carrying out standardization processing on the forward matrix to eliminate the influence among different dimensions.
Assuming n users and m subdivision indexes, the forward matrix is as follows:
let the normalization matrix be denoted as Z, and the value of each entry be:
if there are negative numbers in the normalized matrix, another normalization method is needed for the forward matrix X, which is expressed as:
this allows the normalized matrix to have each value between [0,1 ].
And S3, calculating a probability matrix, namely the proportion of the ith user sample under the jth index.
And calculating the information entropy of each index, calculating the information utility value, and normalizing to obtain the entropy weight of each subdivision index.
Specifically, if the events X are likely to occur as X1, X2.. And xn, respectively, then we define the entropy of the information of the events X, i.e. the expected value of the information amount is:
For the j index, its information entropy is:
the information utility value is then:
d j =1-e j
then, the information utility value is normalized, so that the entropy weight of the ith index can be obtained:
and S4, calculating the score of each user and carrying out normalization processing.
Assuming n users, the normalized matrix of m segment indices is as follows:
wherein z is ij The jth index representing the ith user.
And calculating the distance between the user and the maximum value and the minimum value, calculating the score and carrying out normalization processing.
Wherein, the calculation formula of the maximum value is as follows:
the minimum is calculated as:
thus, the maximum value of the j-th segment index can be calculated based on the above calculation formulas of the maximum value and the minimum valueValue ofAnd a minimum value->
Further, the distance of the ith (i =1,2,3.. The., n) user from the maximum, i.e., the maximum scoring distance of the ith user, may be calculated:
and the distance between the ith user and the minimum value, namely the minimum scoring distance of the ith user:
thus, the score of the ith user without normalization processing can be derived:
Finally, the score is normalized, namely the score s of the user is ordered i Dividing the sum of all the user score values to obtain the final score of the ith user, namely the purchasing power score value or the activity score value of the user.
FIG. 6 schematically shows a flow chart of a method for user population partitioning based on the improved user clustering model according to an embodiment of the present disclosure.
As shown in fig. 6, the method for user group division based on the improved user group division model of this embodiment includes operations S610 to S630.
In operation S610, a user clustering coordinate system is constructed based on the composite index, and the number of dimensions of the user clustering coordinate system is the same as the number of the composite index.
In operation S620, an average index value of each composite index in the improved user clustering model is obtained based on the user transaction data.
In operation S630, quadrant division is performed on the user clustering coordinate system based on the average index value of each of the composite indexes, so as to obtain user subdivision clusters.
For example, in a preferred embodiment of the present disclosure, the composite indicators include a user churn probability, a user liveness score, and a user purchasing power score. Based on the three types of indexes, a three-dimensional coordinate system can be constructed. And each quadrant can be divided into two parts based on the average index value of each comprehensive index, so that the users are divided into eight subdivided groups including important value customers, general value customers, important users to be developed, general customers to be developed, important customers to be maintained, general customers to be maintained, important customers to be restored and general customers to be restored. Thus, different operation strategies can be formulated for different segment groups, as shown in table 1.
TABLE 1
User categories | R value | F value | Value of M | Operation strategy |
Customer of important value | Height of | High (a) | Height of | Keeping the current situation |
General value customer | Height of | High (a) | Is low in | Stimulating consumption |
Important to-be-developed user | Height of | Is low in | Height of | Frequency up |
General customer to be developed | Height of | Is low in | Is low in | Excavation needs |
Customer of importance to be maintained | Is low in | Height of | Height of | User reflow |
General customer to be maintained | Is low in | Height of | Is low in | Loss recall |
(continuation table 1)
Important customer to be retrieved | Is low with | Is low in | Height of | Key recall |
General customer to be retrieved | Is low in | Is low with | Is low in | Take no action |
Based on the construction method of the user marketing strategy, the embodiment of the disclosure also provides a construction device of the user marketing strategy. The apparatus will be described in detail below with reference to fig. 7.
Fig. 7 is a block diagram schematically illustrating a construction apparatus of a user marketing strategy according to an embodiment of the present disclosure.
As shown in fig. 7, the construction apparatus 700 of the user marketing strategy of this embodiment includes an acquisition module 710, a first construction module 720, a second construction module 730, a division module 740, and a strategy construction module 750.
Wherein the obtaining module 710 is configured to obtain user transaction data.
The first construction module 720 is configured to construct a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing transaction conditions and transaction capabilities of the user within a preset time range.
The second construction module 730 is configured to build an improved user clustering model based on the user portrait index system, the improved user clustering model being constructed based on an RFM model, the improved user clustering model comprising a plurality of synthetic indexes, wherein each synthetic index comprises the same or different number of segment indexes.
The partitioning module 740 is configured to perform user population partitioning based on the improved user clustering model.
The policy construction module 750 is configured to construct a user classification operation policy based on the user population division result.
Fig. 8 schematically shows a block diagram of a first building block according to an embodiment of the present disclosure.
As shown in fig. 8, the first building module 720 of this embodiment includes a pre-processing submodule 7201 and a screening submodule 7202.
Wherein the pre-processing submodule 7201 is configured to pre-process the user transaction data, the pre-processing comprising: screening, cleaning, structuring, statistics, and outlier processing.
The screening submodule 7202 is configured to sort and screen the preprocessed user transaction data to construct the user portrait index system.
FIG. 9 schematically shows a block diagram of a policy building module according to an embodiment of the present disclosure.
As shown in fig. 9, the policy building module 750 of this embodiment includes a coordinate system establishing sub-module 7501, a first calculating sub-module 7502, and a second calculating sub-module 7503.
The coordinate system establishing submodule 7501 is configured to establish a user clustering coordinate system based on the comprehensive index, and the dimension of the user clustering coordinate system is the same as the number of the comprehensive indexes.
The first calculation submodule 7502 is configured to obtain an average metric value of each composite metric in the improved user clustering model based on the user transaction data.
The second calculating submodule 7503 is configured to perform quadrant division on the user grouping coordinate system based on the average index value of each comprehensive index, and obtain a user subdivision group.
Fig. 10 schematically shows a block diagram of a second building block according to an embodiment of the present disclosure.
As shown in fig. 10, the second construction module 730 of this embodiment includes a forward normalization sub-module 7301, a normalization sub-module 7302, an entropy weight calculation sub-module 7303, and a score value calculation sub-module 7304.
Wherein the forward sub-module 7301 is configured to obtain a user segment index value, and construct a forward matrix based on the user segment index value, wherein the user segment index value includes m segment index values corresponding to n users, and the forward matrix corresponds to a user purchasing power score or a user activity score.
The entropy weight calculation sub-module 7303 is configured to calculate the information entropy and the information utility value of each subdivision index based on the normalization matrix, and obtain the entropy weight of each subdivision index.
The score value calculation sub-module 7304 is configured to calculate a user purchasing power score value and/or a user activity score value based on the normalization matrix and the entropy weight of each segment index.
According to the embodiment of the disclosure, any multiple modules of the obtaining module 710, the first constructing module 720, the second constructing module 730, the dividing module 740, the policy constructing module 750, the preprocessing sub-module 7201, the screening sub-module 7202, the coordinate system establishing sub-module 7501, the first calculating sub-module 7502, the second calculating sub-module 7503, the forward sub-module 7301, the normalizing sub-module 7302, the entropy weight calculating sub-module 7303 and the score value calculating sub-module 7304 may be combined and implemented in one module, or any one module thereof may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of the other modules and implemented in one module. According to the embodiment of the present disclosure, at least one of the obtaining module 710, the first constructing module 720, the second constructing module 730, the dividing module 740, the policy constructing module 750, the preprocessing sub-module 7201, the screening sub-module 7202, the coordinate system establishing sub-module 7501, the first calculating sub-module 7502, the second calculating sub-module 7503, the normalizing sub-module 7301, the normalizing sub-module 7302, the entropy weight calculating sub-module 7303, and the score value calculating sub-module 7304 may be at least partially implemented as a hardware circuit, such as a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), a system on a chip, a system on a substrate, a system on a package, an Application Specific Integrated Circuit (ASIC), or may be implemented by hardware or firmware such as any other reasonable manner of integrating or packaging the circuit, or implemented by any one of three manners of software, hardware, and firmware, or an appropriate combination of any of them. Alternatively, at least one of the obtaining module 710, the first constructing module 720, the second constructing module 730, the dividing module 740, the policy constructing module 750, the preprocessing sub-module 7201, the screening sub-module 7202, the coordinate system establishing sub-module 7501, the first calculating sub-module 7502, the second calculating sub-module 7503, the forward sub-module 7301, the normalizing sub-module 7302, the entropy weight calculating sub-module 7303, and the score value calculating sub-module 7304 may be at least partially implemented as a computer program module, and when the computer program module is executed, corresponding functions may be executed.
FIG. 11 schematically illustrates a block diagram of an electronic device suitable for implementing a construction method of a user marketing strategy according to an embodiment of the present disclosure.
As shown in fig. 11, an electronic apparatus 900 according to an embodiment of the present disclosure includes a processor 901 which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 902 or a program loaded from a storage portion 908 into a Random Access Memory (RAM) 903. Processor 901 may comprise, for example, a general purpose microprocessor (e.g., a CPU), an instruction set processor and/or associated chipset, and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor 901 may also include on-board memory for caching purposes. The processor 901 may comprise a single processing unit or a plurality of processing units for performing the different actions of the method flows according to embodiments of the present disclosure.
In the RAM 903, various programs and data necessary for the operation of the electronic apparatus 900 are stored. The processor 901, ROM 902, and RAM 903 are connected to each other by a bus 904. The processor 901 performs various operations of the method flows according to the embodiments of the present disclosure by executing programs in the ROM 902 and/or the RAM 903. Note that the programs may also be stored in one or more memories other than the ROM 902 and the RAM 903. The processor 901 may also perform various operations of the method flows according to embodiments of the present disclosure by executing programs stored in the one or more memories.
The present disclosure also provides a computer-readable storage medium, which may be embodied in the device/apparatus/system described in the above embodiments; or may exist separately and not be assembled into the device/apparatus/system. The computer-readable storage medium carries one or more programs which, when executed, implement the method according to an embodiment of the disclosure.
According to embodiments of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example but is not limited to: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. For example, according to embodiments of the present disclosure, a computer-readable storage medium may include the ROM 902 and/or the RAM 903 described above and/or one or more memories other than the ROM 902 and the RAM 903.
Embodiments of the present disclosure also include a computer program product comprising a computer program containing program code for performing the method illustrated in the flow chart. When the computer program product runs in a computer system, the program code is used for causing the computer system to realize the method provided by the embodiment of the disclosure.
The computer program performs the above-described functions defined in the system/apparatus of the embodiments of the present disclosure when executed by the processor 901. The above described systems, devices, modules, units, etc. may be implemented by computer program modules according to embodiments of the present disclosure.
In one embodiment, the computer program may be hosted on a tangible storage medium such as an optical storage device, a magnetic storage device, or the like. In another embodiment, the computer program may also be transmitted in the form of a signal over a network medium, distributed, and downloaded and installed via the communication section 909 and/or installed from the removable medium 911. The computer program containing program code may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the foregoing.
In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 909, and/or installed from the removable medium 911. The computer program, when executed by the processor 901, performs the above-described functions defined in the system of the embodiment of the present disclosure. The systems, devices, apparatuses, modules, units, etc. described above may be implemented by computer program modules according to embodiments of the present disclosure.
In accordance with embodiments of the present disclosure, program code for executing computer programs provided by embodiments of the present disclosure may be written in any combination of one or more programming languages, and in particular, these computer programs may be implemented using high level procedural and/or object oriented programming languages, and/or assembly/machine languages. The programming language includes, but is not limited to, programming languages such as Java, C + +, python, the "C" language, or the like. The program code may execute entirely on the user computing device, partly on the user device, partly on a remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computing device (e.g., through the internet using an internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Those skilled in the art will appreciate that various combinations and/or combinations of features recited in the various embodiments and/or claims of the present disclosure can be made, even if such combinations or combinations are not expressly recited in the present disclosure. In particular, various combinations and/or combinations of the features recited in the various embodiments and/or claims of the present disclosure may be made without departing from the spirit or teaching of the present disclosure. All such combinations and/or associations are within the scope of the present disclosure.
The embodiments of the present disclosure have been described above. However, these examples are for illustrative purposes only and are not intended to limit the scope of the present disclosure. Although the embodiments are described separately above, this does not mean that the measures in the embodiments cannot be used in advantageous combination. The scope of the disclosure is defined by the appended claims and equivalents thereof. Various alternatives and modifications can be devised by those skilled in the art without departing from the scope of the present disclosure, and such alternatives and modifications are intended to be within the scope of the present disclosure.
Claims (13)
1. A method for constructing a user marketing strategy is characterized by comprising the following steps:
acquiring user transaction data;
constructing a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing the transaction condition and the transaction capacity of a user within a preset time range;
establishing an improved user clustering model based on the user portrait index system, wherein the improved user clustering model is established based on an RFM (remote object model), and comprises a plurality of comprehensive indexes, wherein each comprehensive index comprises the same or different numbers of subdivision indexes;
dividing user groups based on the improved user group dividing model; and
and constructing a user classification operation strategy based on the user group division result.
2. The method of claim 1, wherein the user representation index system includes a plurality of user representation dimensions, wherein each user representation dimension includes a plurality of user transaction data types, the composite index is built based on a combination of the user transaction data types, and the segment index is associated with the user transaction data types.
3. The method of claim 2, wherein the user representation dimensions include user transaction habits, user transaction preferences, and user value, wherein the user transaction habits include user transaction frequency data and user transaction intent data; and/or the user transaction preference comprises the interest degree of a user product and the acceptance degree of a user marketing activity; and/or the user value includes user asset data and user transaction amount data.
4. The method of claim 3, wherein the user transaction frequency data includes a number of user continuous transaction days and a number of user transactions within a year, and the user transaction intent data includes specific gravity data of user monthly transaction amount and user monthly income;
and/or, the user asset data comprises user annual revenue; and
and/or the user transaction amount data comprises a user annual transaction amount and a user historical average transaction amount.
5. The method of claim 1 or 2, wherein the composite indicators comprise a user purchasing power score, a user activity score, and a user churn probability, wherein the segment indicators in the user purchasing power score comprise a user annual transaction amount, a user total transaction amount, and a user annual income;
and/or the segment indexes in the activity degree score comprise the annual transaction times of the user, the average transaction interval days of the user and the continuous transaction days of the user;
and/or the segment indexes in the attrition probability comprise the number of continuous transaction days of the user, the number of transactions within one year of the user, the interest degree of the user product and the acceptance degree of the user marketing activity.
6. The method of claim 5, wherein the user purchasing power score and the user liveness score are calculated based on a Topsis method in combination with an entropy weight method;
and/or the user churn probability is calculated based on a classification algorithm, wherein the user purchasing power score and the user activity score are calculated based on a Topsis method combined with an entropy weight method, and the calculation comprises the following steps:
acquiring user subdivision index values, and constructing a forward matrix based on the user subdivision index values, wherein the user subdivision index values comprise m subdivision index values corresponding to n users, and the forward matrix corresponds to user purchasing power scores or user activity scores;
carrying out standardization processing on the forward matrix to obtain a standardized matrix;
calculating the information entropy and the information utility value of each subdivision index based on the standardized matrix, and acquiring the entropy weight of each subdivision index; and
and calculating the purchasing power scoring value and/or the activity scoring value of the user based on the standardized matrix and the entropy weight of each subdivision index.
7. The method of claim 6, wherein calculating a user buying power score value and/or a user liveness score value based on the normalization matrix and the entropy weight of each segment indicator comprises:
respectively calculating the jth subdivision index maximum value and the jth subdivision index minimum value based on the standardized matrix;
calculating the maximum scoring distance of the ith user based on the normalization matrix, the entropy weight of each subdivision index and the maximum value of each subdivision index;
calculating the minimum scoring distance of the ith user based on the normalization matrix, the entropy weight of each subdivision index and the minimum value of each subdivision index; and
and calculating a user score value of the ith user based on the maximum scoring distance of the ith user and the minimum scoring distance of the ith user, wherein the user score value is a user purchasing power score value or a user activity score value.
8. The method of claim 1, wherein the user population partitioning based on the improved user clustering model comprises:
constructing a user grouping coordinate system based on the comprehensive indexes, wherein the dimension of the user grouping coordinate system is the same as the number of the comprehensive indexes;
acquiring the average index value of each comprehensive index in the improved user clustering model based on user transaction data; and
and performing quadrant division on the user grouping coordinate system based on the average index value of each comprehensive index to obtain user grouping groups.
9. The method of claim 1, wherein said constructing a user representation index system based on said user transaction data comprises:
pre-processing the user transaction data, the pre-processing comprising: screening, cleaning, structuring, counting and abnormal value processing; and
and classifying and screening the preprocessed user transaction data to construct the user portrait index system.
10. An apparatus for constructing a marketing strategy for a user, comprising:
an acquisition module configured to acquire user transaction data;
the first construction module is configured to construct a user portrait index system based on the user transaction data, wherein the user portrait index system is used for representing the transaction condition and the transaction capability of a user within a preset time range;
the second construction module is configured to establish an improved user clustering model based on the user portrait index system, the improved user clustering model is constructed based on an RFM model, and the improved user clustering model comprises a plurality of comprehensive indexes, wherein each comprehensive index comprises the same or different numbers of subdivision indexes;
a dividing module configured to divide a user group based on the improved user grouping model; and
and the strategy construction module is configured to construct a user classification operation strategy based on the user group division result.
11. An electronic device, comprising:
one or more processors;
a storage device to store one or more programs,
wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to perform the method of any of claims 1-9.
12. A computer readable storage medium having stored thereon executable instructions which, when executed by a processor, cause the processor to perform the method according to any one of claims 1 to 9.
13. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310042183.2A CN115983900A (en) | 2023-01-12 | 2023-01-12 | Method, apparatus, device, medium, and program product for constructing user marketing strategy |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310042183.2A CN115983900A (en) | 2023-01-12 | 2023-01-12 | Method, apparatus, device, medium, and program product for constructing user marketing strategy |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115983900A true CN115983900A (en) | 2023-04-18 |
Family
ID=85959595
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310042183.2A Pending CN115983900A (en) | 2023-01-12 | 2023-01-12 | Method, apparatus, device, medium, and program product for constructing user marketing strategy |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115983900A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116595342A (en) * | 2023-07-07 | 2023-08-15 | 北京数巅科技有限公司 | Crowd circling method, device and equipment and storage medium |
CN117094722A (en) * | 2023-10-19 | 2023-11-21 | 深圳薪汇科技有限公司 | Security supervision method and system for online payment |
-
2023
- 2023-01-12 CN CN202310042183.2A patent/CN115983900A/en active Pending
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116595342A (en) * | 2023-07-07 | 2023-08-15 | 北京数巅科技有限公司 | Crowd circling method, device and equipment and storage medium |
CN116595342B (en) * | 2023-07-07 | 2023-09-29 | 北京数巅科技有限公司 | Crowd circling method, device and equipment and storage medium |
CN117094722A (en) * | 2023-10-19 | 2023-11-21 | 深圳薪汇科技有限公司 | Security supervision method and system for online payment |
CN117094722B (en) * | 2023-10-19 | 2024-01-30 | 深圳薪汇科技有限公司 | Security supervision method and system for online payment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101939554B1 (en) | Determining a temporary transaction limit | |
CN108133418A (en) | Real-time credit risk management system | |
CN115983900A (en) | Method, apparatus, device, medium, and program product for constructing user marketing strategy | |
CN110705719A (en) | Method and apparatus for performing automatic machine learning | |
US20210125272A1 (en) | Using Inferred Attributes as an Insight into Banking Customer Behavior | |
Boz et al. | Reassessment and monitoring of loan applications with machine learning | |
WO2023018976A2 (en) | Methods and computer systems for automated event detection based on machine learning | |
CN115545886A (en) | Overdue risk identification method, overdue risk identification device, overdue risk identification equipment and storage medium | |
Balakayeva et al. | The solution to the problem of processing Big Data using the example of assessing the solvency of borrowers | |
Zhu et al. | A DEALG methodology for prediction of effective customers of internet financial loan products | |
Yang et al. | Is there the Pareto principle in public library circulation? A case study of one public library in Taiwan | |
CN115204881A (en) | Data processing method, device, equipment and storage medium | |
Li et al. | An improved genetic-XGBoost classifier for customer consumption behavior prediction | |
CN115641198A (en) | User operation method, device, electronic equipment and storage medium | |
CN114693409A (en) | Product matching method, device, computer equipment, storage medium and program product | |
CN112288564B (en) | Method and system for generating credit level of agricultural social service main body | |
Bello | The Impact of Big Data on Economic Forecasting and Policy Making | |
CN117035843A (en) | Customer loss prediction method and device, electronic equipment and medium | |
TWI657393B (en) | Marketing customer group prediction system and method | |
Lohse | Machine Learning in Banking: Exploring the feasibility of using consumer level bank transaction data for credit risk evaluation | |
CN118154254A (en) | Customer group production method, device, electronic equipment and medium | |
CN115860815A (en) | Merchant preference distribution method, device, equipment and medium | |
CN118229427A (en) | Risk group identification method, apparatus, electronic device, medium and program product | |
CN116562974A (en) | Object recognition method, device, electronic equipment and storage medium | |
CN118195642A (en) | Information processing method, apparatus, device, storage medium and program product |
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 |