CN115170294A

CN115170294A - Client classification method and device and server

Info

Publication number: CN115170294A
Application number: CN202210866392.4A
Authority: CN
Inventors: 吴政楠; 林慕云; 章宗杰; 张志雄
Original assignee: Industrial and Commercial Bank of China Ltd ICBC
Current assignee: Industrial and Commercial Bank of China Ltd ICBC
Priority date: 2022-07-22
Filing date: 2022-07-22
Publication date: 2022-10-11

Abstract

The specification provides a client classification method, a client classification device and a server. The method comprises the following steps: determining a customer type for the customer; determining an initial clustering center according to the customer type; acquiring attribute data and behavior data of a client; constructing a value vector of the client by utilizing the attribute data and the behavior data of the client according to a preset processing rule; and according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer. The method based on the description can well realize clustering processing aiming at the clients and obtain a classification result which is more accurate and has higher reference value; and according to the classification result, the matched promotion data is pushed to the client in a targeted manner, so that a better pushing effect is obtained.

Description

Client classification method, device and server

Technical Field

The specification belongs to the field of big data, and particularly relates to a client classification method, a client classification device and a server.

Background

In general, a platform organization such as a bank tends to hold a huge amount of customer data. Based on business requirements, the platform mechanism needs to classify the customers based on big data algorithm by using the customer data.

Based on the existing method, the customers are often classified by using a clustering algorithm to obtain a corresponding classification result. However, the classification result obtained based on the existing method is often poor in accuracy and low in reference value.

In view of the above technical problems, no effective solution has been proposed at present.

Disclosure of Invention

The specification provides a client classification method, a client classification device and a server, which can solve the problem that the classification precision in the existing method is not ideal enough, and realize accurate classification of clients.

An object of an embodiment of the present specification is to provide a customer classification method, including:

determining a customer type for the customer; determining an initial clustering center according to the customer type;

acquiring attribute data and behavior data of a client;

constructing a value vector of the client by utilizing the attribute data and the behavior data of the client according to a preset processing rule;

and according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer.

Further, in another embodiment of the method, the client type of the client includes: the client system comprises clients with high current value, high potential value and high dependency, clients with high current value, high potential value and low dependency, clients with high current value, low potential value and high dependency, clients with high current value, low potential value and low dependency, clients with low current value, high potential value and high dependency, clients with low current value, high potential value and low dependency, clients with low current value, low potential value and high dependency, and clients with low current value, low potential value and high dependency.

Further, in another embodiment of the method, the constructing a value vector of the customer by using the attribute data and the behavior data of the customer according to a preset processing rule includes:

according to a preset processing rule, determining the current value characteristic, the potential value characteristic and the dependency characteristic of the client by utilizing the attribute data and the behavior data of the client respectively;

and combining the current value characteristics, the potential value characteristics and the dependency characteristics of the client to obtain a value vector of the client.

Further, in another embodiment of the method, the determining the current value characteristic of the customer by using the attribute data and the behavior data of the customer according to the preset processing rule includes:

obtaining a latest transaction time equal interval, a transaction frequency equal interval and a total transaction amount equal interval according to latest transaction time data, transaction frequency data and total transaction amount data in the behavior data of the client;

quantifying the latest transaction time data, the transaction frequency data and the total transaction amount data according to the latest transaction time equal interval, the transaction frequency equal interval and the total transaction amount equal interval to obtain a target latest transaction time value, a target transaction frequency value and a target total transaction amount value;

determining weights for the target recent transaction time value, the target transaction frequency value, and the target total transaction amount value based on an analytic hierarchy process;

and determining the current value characteristic of the customer according to the target recent transaction time value, the target transaction frequency value, the weight of the target total transaction amount value, the target recent transaction time value, the target transaction frequency value and the target total transaction amount value.

Further, in another embodiment of the method, the determining the potential value characteristics of the customer by using the attribute data and the behavior data of the customer according to the preset processing rule includes:

according to the age data, the gender data, the education background data and the income level data of the customers in the attribute data of the customers, calculating a quantized age value, a quantified gender value, a quantified education background value and a quantified income level value;

and determining potential value characteristics of the customers according to the quantified age numerical value, gender numerical value, education background numerical value and income level numerical value.

Further, in another embodiment of the method, the obtaining a classification result of the customer by performing a preset clustering process on the customer according to the initial clustering center and the value vector of the customer includes:

and according to the initial clustering center and the value vector of the client, carrying out preset clustering processing on the client by using an improved K-means algorithm to obtain a classification result of the client.

Further, in another embodiment of the method, the obtaining the classification result of the customer by performing a preset clustering process on the customer by using the improved K-means algorithm includes:

the current clustering processing is carried out on the clients by using the improved K-means algorithm according to the following modes:

acquiring a last clustering center and a last classification result;

calculating the current clustering center according to the last clustering center and the last classification result;

and according to the current clustering center and the value vector of the client, clustering the client to obtain the current classification result of the client.

Further, in another embodiment of the method, after obtaining the classification result of the customer by performing a preset clustering process on the customer according to the initial clustering center and the value vector of the customer, the method further includes:

determining the clients contained in different client type groups according to the classification result of the clients;

configuring matched promotion data according to different client types;

and pushing matched promotion data to the clients contained in the different client type groups.

In another aspect, the present application provides a customer classification device, including:

the center generation module is used for determining the type of the client aiming at the client; determining an initial clustering center according to the customer type;

the acquisition module is used for acquiring attribute data and behavior data of the client;

the construction module is used for constructing a value vector of the client by utilizing the attribute data and the behavior data of the client according to a preset processing rule;

and the classification module is used for carrying out preset clustering processing on the customers according to the initial clustering centers and the value vectors of the customers so as to obtain the classification results of the customers.

In another aspect, the present application provides a server comprising a processor and a memory for storing processor-executable instructions, which when executed by the processor implement the steps of any one of the method embodiments in the embodiments of the present specification.

In yet another aspect, the present application provides a computer-readable storage medium having stored thereon computer instructions which, when executed, implement the steps of any one of the method embodiments in the embodiments of the present specification.

The method, the device and the server for classifying the clients provided by the specification determine the types of the clients; determining an initial clustering center according to the customer type; acquiring attribute data and behavior data of a client; constructing a value vector of the customer by using the attribute data and the behavior data of the customer according to a preset processing rule; and according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer. Before specific clustering, determining a required client type by combining a specific service scene, and determining an initial clustering center with relatively good effect to participate in subsequent clustering processing according to the client type; during specific clustering, the clustering center obtained last time is considered and used, and the improved K-means algorithm is utilized to perform the current preset clustering treatment, so that the classification for customers can be better realized, and the classification result with higher accuracy and higher reference value is obtained; and according to the classification result, the matched promotion data is pushed to the client in a targeted manner, so that a better pushing effect is obtained.

In addition, when the value vector of the client is specifically constructed by using the attribute data and the behavior data of the client according to the preset processing rule, the current value characteristic, the potential value characteristic and the dependency characteristic of the client can be respectively determined by using the attribute data and the behavior data of the client according to the preset processing rule; and combining the current value characteristic, the potential value characteristic and the dependency characteristic of the client, so that the value vector of the client with a good clustering effect can be obtained.

Drawings

In order to more clearly illustrate the embodiments of the present specification, the drawings needed to be used in the embodiments will be briefly described below, and the drawings in the following description are only some of the embodiments described in the specification, and it is obvious to those skilled in the art that other drawings can be obtained based on the drawings without any inventive work.

FIG. 1 is a schematic flow chart diagram illustrating one embodiment of a customer categorization method provided herein;

FIG. 2 is a flow chart illustrating a customer classification method in an example of a particular scenario provided herein;

FIG. 3 is a block diagram of an embodiment of a customer sorting apparatus provided herein;

fig. 4 is a block diagram of a hardware configuration of an embodiment of a server provided by the present specification.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in the present specification, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only a part of the embodiments of the present specification, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present specification without making any creative effort shall fall within the protection scope of the present specification.

Considering that the traditional K-means clustering method is a common classification method in the field of big data, the method randomly generates an initial clustering center, only considers the influence of the previous clustering result on the current clustering center in the clustering calculation process, and does not consider the influence of the previous clustering center on the current clustering center, so that the client classification result obtained by calculation is not accurate enough.

Furthermore, it is also considered that in the existing bank customer classification method, only the current value characteristics of the customer are generally considered, but the potential value characteristics of the customer are not considered, which results in that service personnel of the bank cannot push promotion data meeting the requirements of the customer with the potential value aiming at the customer with the potential value, and further leads the bank to lose the customer with the potential value while paying the manpower service cost.

Aiming at the problems existing in the existing method and the specific reasons for the problems, the application considers the introduction of a customer classification method based on an improved K-means algorithm to realize accurate classification of customers.

Based on the above thought, the present specification proposes a customer classification method, first, determining a customer type for a customer; determining an initial clustering center according to the customer type; acquiring attribute data and behavior data of a client; secondly, according to a preset processing rule, utilizing the attribute data and the behavior data of the client to construct a value vector of the client; and finally, according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer.

Referring to fig. 1, an embodiment of the present disclosure provides a method for classifying customers. In particular implementations, the method may include the following.

S101: determining a customer type for the customer; and determining an initial clustering center according to the customer type.

In some embodiments, taking a transaction scenario faced by a platform mechanism such as a bank as an example, a historical transaction record may be obtained and utilized, and a specific business requirement is combined, and a big data analysis is performed based on three dimensions of a current value, a potential value, and a dependency, so as to determine a required customer type, which may specifically include: the client system comprises clients with high current value, high potential value and high dependency, clients with high current value, high potential value and low dependency, clients with high current value, low potential value and high dependency, clients with high current value, low potential value and low dependency, clients with low current value, high potential value and high dependency, clients with low current value, high potential value and low dependency, clients with low current value, low potential value and high dependency, and clients with low current value, low potential value and high dependency.

The customers with high current value, high potential value and high dependency degree can be specifically understood as customers with current value characteristics meeting preset current value requirements, potential value characteristics meeting preset potential value requirements, and dependency degree characteristics meeting preset dependency degree requirements. For example, a gold premium customer, which is characterized by having completed an overabundance of transactions at the current bank, has a high current value; the potential future brings a great amount of profits to the current bank, so the potential future has high value; meanwhile, the customers have high loyalty to the current bank and low loss risk, so the customers have high dependence.

The customers with high current value, high potential value and low dependency degree can be specifically understood as customers with current value characteristics meeting preset current value requirements, potential value characteristics meeting preset potential value requirements, and dependency degree characteristics not meeting preset dependency degree requirements. For example, a premium customer to be retained, which is characterized by having completed an excessive transaction at the current bank, and therefore has a high current value; the potential future brings a great amount of profits to the current bank, so the potential value is high; meanwhile, the loyalty of the customers to the current bank is low, and the customers may be diverted to other banks in the future, so that the customers need to be retained in a proper mode.

The customers with high current value, low potential value and high dependency degree can be specifically understood as customers with current value characteristics meeting preset current value requirements, potential value characteristics not meeting preset potential value requirements, and dependency degree characteristics meeting preset dependency degree requirements. For example, general customers with undesirable future value are characterized by having completed an excessive amount of transactions at the current bank and therefore have a high current value; but it is difficult to bring a lot of profits to the current bank in the future, so the method has low potential value; meanwhile, the customer has high loyalty to the current bank and low loss risk, so the customer has high dependence.

The customers with high current value, low potential value and low dependency can be specifically understood as customers whose current value characteristics meet preset current value requirements, whose potential value characteristics do not meet preset potential value requirements, and whose dependency characteristics do not meet preset dependency requirements. For example, an impotent customer who will be lost is characterized by having completed an overabundance of transactions at the current bank and therefore has a high current value; but the potential value is low; meanwhile, the loyalty of the customers to the current bank is low, so that the dependence is low; therefore, such customers are easily lost and cannot bring huge profits to the current bank in the future without being saved.

The customers with low current value, high potential value and high dependency can be specifically understood as customers with current value characteristics meeting non-preset current value requirements, potential value characteristics meeting preset potential value requirements and dependency characteristics meeting preset dependency requirements. For example, a potential premium customer, which is characterized by not having an excessive amount of transactions at the current bank, therefore has a low current value; the potential future brings a great amount of profits to the current bank, so the potential value is high; meanwhile, the customers have high loyalty to the current bank and low loss risk, so the customers have high dependence.

The customers with low current value, high potential value and low dependency can be specifically understood as customers whose current value characteristics meet the current value requirement which is not preset, whose potential value characteristics meet the preset potential value requirement, and whose dependency characteristics do not meet the preset dependency requirement. For example, there is a need to save potential customers who are characterized by not having an excessive number of transactions at the current bank, and therefore having a low current value; the potential future brings a great amount of profits to the current bank, so the potential future has high value; meanwhile, the loyalty of the customers to the current bank is low, and the customers may be diverted to other banks in the future, so that the customers need to be retained in a proper manner.

The customers with low current value, low potential value and high dependency can be specifically understood as customers whose current value characteristics meet current value requirements which are not preset, whose potential value characteristics do not meet preset potential value requirements, and whose dependency characteristics meet preset dependency requirements. For example, a general value customer, which is characterized by not having an excessive transaction at the current bank, has a low current value; the method has the advantages that large profits are difficult to bring to the current bank in the future, so that the method has low potential value; meanwhile, the customer has high loyalty to the current bank, the loss risk is low, and small amount transaction is often carried out in the current bank, so that the customer has high dependence.

The customers with low current value, low potential value and low dependency degree can be specifically understood as customers whose current value characteristics do not meet the preset current value requirement, whose potential value characteristics do not meet the preset potential value requirement, and whose dependency degree characteristics do not meet the preset dependency degree requirement. For example, a completely worthless customer, characterized by not having an excessive number of transactions at the current bank, and therefore having a low current value; the method has the advantages that large profits are difficult to bring to the current bank in the future, so that the method has low potential value; meanwhile, the loyalty of the customers to the current bank is low, and the dependence is low; therefore, such customers are prone to loss and cannot bring huge profits to the current bank in the future without being saved.

In some embodiments, the preset current value requirement may specifically be: under the condition that the current value characteristic of the client is smaller than a first preset value, the client is considered not to meet the preset current value requirement; and under the condition that the current value characteristic of the client is greater than or equal to the first preset value, the client is considered to meet the preset current value requirement.

In some embodiments, the preset potential value requirement may be specifically: under the condition that the potential value characteristic of the customer is smaller than a second preset value, the customer is considered not to meet the preset potential value requirement; and in the case that the potential value characteristic of the client is greater than or equal to the second preset value, the client is considered to meet the preset potential value requirement.

In some embodiments, the preset dependency requirement may specifically be: under the condition that the dependency characteristics of the client are smaller than a third preset value, the client is considered to be not in accordance with the preset dependency requirement; and in the case that the dependency characteristics of the client are greater than or equal to a third preset value, the client is considered to meet the preset dependency requirement.

In some embodiments, the determining the initial cluster center may include: obtaining and utilizing historical transaction records, screening representative sample clients conforming to various client types, and constructing to obtain a plurality of sample type client groups; determining the current value characteristics, potential value characteristics and dependency characteristics of sample clients contained in each sample type client group; and calculating the average value of the current value characteristics, the average value of the potential value characteristics and the average value of the dependency characteristics of each sample type group according to the current value characteristics, the potential value characteristics and the dependency characteristics of the sample clients contained in each sample type group, and taking the average values as the initial clustering center.

Based on the above embodiments, a plurality of customer types for the customer may be determined, and corresponding initial cluster centers may be generated based on the plurality of customer types.

S102: attribute data and behavior data of the client are acquired.

In some embodiments, the attribute data of the client may specifically include: age data, gender data, educational background data, income level data.

In some embodiments, the behavior data of the client may specifically include: recent transaction time data, transaction frequency data, total transaction amount data, quality evaluation data, service evaluation data, price sensitivity data, convenience evaluation data, store evaluation data, purchase relationship establishment duration data, annual purchase frequency data, annual purchase amount data, and cross-selling ability value data.

In some embodiments, the obtaining of the attribute data of the client may include, when implemented: the attribute data of the client is obtained by querying a user database of the client.

In some embodiments, the obtaining of the behavior data of the client may include, in specific implementation: and acquiring the behavior data of the client by inquiring and according to the historical operation record of the client. For example, by querying and counting the last week of transaction records of a customer, transaction frequency data for the customer is determined.

It should be noted that the data related to the user, which is acquired and used in the present application, is acquired and used on the premise that the user knows and agrees. In addition, the data acquisition, storage, use, processing and the like in the technical scheme of the application all conform to relevant regulations of national laws and regulations.

S103: and constructing a value vector of the customer by using the attribute data and the behavior data of the customer according to a preset processing rule.

In some embodiments, the constructing a value vector of the customer by using the attribute data and the behavior data of the customer according to the preset processing rule may include:

s1: according to a preset processing rule, determining the current value characteristic, the potential value characteristic and the dependency characteristic of the client by utilizing the attribute data and the behavior data of the client respectively;

s2: and combining the current value characteristics, the potential value characteristics and the dependency characteristics of the client to obtain a value vector of the client.

In some embodiments, the dependency characteristics may specifically include: psychological dependence characteristics and behavioral dependence characteristics.

In some embodiments, the psychological dependence characteristics may specifically include: quality evaluation parameters, service evaluation parameters, price sensitivity parameters, convenience evaluation parameters, store evaluation parameters.

In some embodiments, the behavior dependency characteristics may specifically include a purchase duration parameter, a purchase frequency parameter, a purchase amount parameter, and a cross-sale parameter.

In some embodiments, the determining, according to the preset processing rule, the current value characteristic of the customer by using the attribute data and the behavior data of the customer may include:

s1: obtaining a latest transaction time equal interval, a transaction frequency equal interval and a total transaction amount equal interval according to latest transaction time data, transaction frequency data and total transaction amount data in the behavior data of the client;

s2: quantifying the latest transaction time data, the transaction frequency data and the total transaction amount data according to the latest transaction time equal interval, the transaction frequency equal interval and the total transaction amount equal interval to obtain a target latest transaction time value, a target transaction frequency value and a target total transaction amount value;

s3: determining weights for the target recent transaction time value, the target transaction frequency value, the target total transaction dollar value based on an analytic hierarchy process;

s4: determining a current value characteristic of the customer according to the target recent transaction time value, the target transaction frequency value, the weight of the target total transaction fund value, the target recent transaction time value, the target transaction frequency value and the target total transaction fund value.

In some embodiments, the preset processing rule may be specifically an RFM model, which is not limited in this specification.

In some embodiments, the obtaining, according to the latest transaction time data, the transaction frequency data, and the total transaction amount data in the behavior data of the customer, the latest transaction time partition, the transaction frequency partition, and the total transaction amount partition may include:

s1: determining the upper limit and the lower limit of the equal interval of the latest transaction time according to the latest transaction time data in the behavior data of the client; within the upper limit and the lower limit of the recent transaction time equal interval, carrying out five equal intervals according to a preset first division criterion to obtain the recent transaction time equal interval;

s2: determining the upper limit and the lower limit of a transaction frequency equal interval according to transaction frequency data in the behavior data of the client; within the upper limit and the lower limit of the transaction frequency equal partition interval, carrying out five equal partitions according to a preset second partition criterion to obtain the transaction frequency equal partition interval;

s3: determining the upper limit and the lower limit of the equal interval of the total transaction amount according to the total transaction amount data in the behavior data of the client; and within the upper limit and the lower limit of the total transaction amount equal division interval, carrying out five equal division according to a preset third division criterion to obtain the total transaction amount equal division interval.

In some embodiments, the above-mentioned last transaction time equally dividing interval may specifically include five sub-intervals, and each sub-interval has a corresponding target quantization value of the last transaction time data, so as to quantize the last transaction time data meeting the sub-interval value range. For example, by querying the latest transaction time data of all the clients to be classified, determining that the minimum value of the latest transaction time data is 1 day, thus determining that the lower limit of the latest transaction time equally-divided interval is equal to 1, determining that the maximum value of the latest transaction time data is 500 days, thus determining that the upper limit of the latest transaction time equally-divided interval is equal to 500, and determining that the latest transaction time interval is [1,500], then dividing the latest transaction time interval according to a preset first division criterion to obtain the latest transaction time equally-divided interval, and setting a corresponding quantization rule: for the latest transaction time data falling into the subintervals [1,100 ], the quantitative value is 1, and the value is taken as a target latest transaction time value; for the latest transaction time data falling into the subintervals [100,200 ], the quantitative value is 2, and the value is taken as a target latest transaction time value; for the latest transaction time data falling into the subintervals [200, 300), the quantitative value is 3, and the value is taken as the target latest transaction time value; quantifying the latest transaction time data falling into the subinterval [300,400 ] to obtain a value of 4, and taking the value as a target latest transaction time value; for the latest transaction time data falling within the subinterval [400,500], the quantization value is 5, and the value is taken as the target latest transaction time value. The above embodiment takes the latest transaction time equal partition as an example for illustration, and other equal partitions are similar, which is not described in detail in this specification.

In some embodiments, the transaction frequency equal-division interval may specifically include five sub-intervals, and each sub-interval has a corresponding target quantization value of the transaction frequency data, so as to quantize the transaction frequency data that conforms to the sub-interval value range.

In some embodiments, the total transaction amount equally dividing interval may specifically include five sub-intervals, and each sub-interval has a corresponding target quantization value of the total transaction amount data, which is used for quantizing the total transaction amount data that conforms to the sub-interval value range.

In the above embodiments, the embodiments of the present specification are described by taking the RFM model as an example, and the embodiments of the present specification may also use other models to obtain the relevant parameters, which is not limited to this.

In some embodiments, the determining the current value characteristic of the customer according to the weight of the target recent transaction time value, the target transaction frequency value, and the target total transaction amount value, and the weight of the target recent transaction time value, the target transaction frequency value, and the target total transaction amount value may include:

the current value characteristic of the customer is calculated according to the following formula:

x＝w _R ×R+w _F ×F+w _M ×M (1)

where x represents the current value characteristic, R represents the target recent transaction time value, w _R Weight representing the target recent trading value, F represents the target trading frequency value, w _F Weight representing target transaction frequency value, M represents target Total transaction amount value, w _M A weight representing a target total transaction amount value.

In some embodiments, the target recent trading time value, the target trading frequency value, and the target total trading value are defined as shown in table 1.

TABLE 1 target recent transaction time value, target transaction frequency value, target Total transaction amount value meaning

In some embodiments, the determining the potential value characteristics of the customer by using the attribute data and the behavior data of the customer according to the preset processing rule may include:

s1: according to the age data, the gender data, the education background data and the income level data of the clients in the attribute data of the clients, calculating quantized age values, gender values, education background values and income level values;

s2: and determining potential value characteristics of the customers according to the quantified age numerical value, gender numerical value, education background numerical value and income level numerical value.

In some embodiments, the age data, the gender data, the education background data, and the income level data of the customer in the attribute data of the customer are calculated, and the quantified age value, the quantified gender value, the quantified education background value, and the income level value may be quantified by referring to the quantification rule set in table 2, or a corresponding quantification rule may be set according to an actual scene, which is not limited in this specification.

In some embodiments, the above-mentioned educational background data may specifically refer to a scholarly calendar of the customer; the income level data may specifically refer to a monthly income amount of the customer.

TABLE 2 age data, gender data, educational background data, income level data quantification rules

In some embodiments, the determining the potential value characteristics of the customer according to the quantified age value, gender value, education background value and income level value may include:

s1: acquiring the weights of the quantified age numerical value, gender numerical value, education background numerical value and income level numerical value based on an analytic hierarchy process;

s2: and determining the potential value characteristics of the customers according to the quantified age value, gender value, education background value and income level value and the weights of the quantified age value, gender value, education background value and income level value.

It should be noted that the present specification may also use other methods to determine the weight of the quantified age value, gender value, educational background value, and income level value, which is not limited herein.

In some embodiments, the determining the dependency characteristics of the client by using the attribute data and the behavior data of the client according to the preset processing rule may include:

s1: quantifying a quality evaluation parameter, a service evaluation parameter, a price sensitivity parameter, a convenience evaluation parameter and a shop evaluation parameter in the psychological dependence characteristics according to a quantification rule set in table 3 by using quality evaluation data, service evaluation data, price sensitivity data, convenience evaluation data and shop evaluation data in name data and behavior data in the attribute data of the client, and calculating a target psychological dependence characteristic according to a quantification result and a weight corresponding to the quantification result;

s2: establishing duration data, annual purchase frequency data, annual purchase quantity data and cross sales capacity data by utilizing the name data in the attribute data of the client and the purchase relation in the behavior data, quantizing a purchase duration parameter, a purchase frequency parameter, a purchase quantity parameter and a cross sales capacity data in the behavior dependency characteristics according to a quantization rule set in a table 4, and calculating target behavior dependency characteristics according to a quantization result and the weight corresponding to the quantization result;

s3: and determining the dependency characteristics of the client according to the target psychological dependency characteristics, the target behavior dependency characteristics, the weight of the preset target psychological dependency characteristics and the weight of the preset target behavior dependency characteristics.

In some embodiments, the quality evaluation data may specifically include 5 grades of poor, general, better, good, and good. The above embodiment takes the quality evaluation data as an example for illustration, and other data are similar and are not described again.

In some embodiments, the convenience evaluation parameter is used for representing the convenience degree of the customer for purchasing the bank goods; the shop evaluation parameters are used for representing the evaluation level of the customer on the bank product brand image shop.

TABLE 3 parameters and weight acquisition rules in psychological dependence characteristics

TABLE 4 parameters and weight acquisition rules in behavior dependency characteristics

In some embodiments, the weight of the preset target psychological dependency characteristic may be specifically 0.75, the weight of the preset target behavioral dependency characteristic may be specifically 0.25, and the embodiment of the present disclosure may also set the weights in other manners, which is not limited thereto.

In some embodiments, the combining the current value feature, the potential value feature, and the dependency feature of the customer to obtain a value vector of the customer may include:

the value vector of the customer is obtained according to the following formula:

value＝(x,y,z) (2)

wherein value represents a value vector of the customer, x represents a current value feature, y represents a potential value feature, and z represents a dependency feature.

Based on the embodiment, the influence of the current value characteristic, the potential value characteristic and the dependency characteristic of the client on the subsequent classification result is considered, and the rationality and the accuracy of the classification result are improved.

S104: and according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer.

In some embodiments, the obtaining a classification result of the customer by performing a preset clustering process on the customer according to the initial clustering center and the value vector of the customer may include: and according to the initial clustering center and the value vector of the client, carrying out preset clustering processing on the client by using an improved K-means algorithm to obtain a classification result of the client.

In some embodiments, the above-mentioned using the improved K-means algorithm to perform a preset clustering process on the customer to obtain a classification result of the customer may be implemented by:

s1: acquiring a last clustering center and a last classification result;

s2: calculating the current clustering center according to the last clustering center and the last classification result;

s3: and according to the current clustering center and the value vector of the client, clustering the client to obtain the current classification result of the client.

In some embodiments, the calculating a current clustering center according to a previous clustering center and a previous classification result may include: and calculating the average value of the value vectors of each client in the last clustering center and the classification result corresponding to the last clustering center, and taking the average value as the current clustering center.

In some embodiments, the clustering, according to the current clustering center and the value vector of the customer, the current classification result of the customer is obtained by clustering the customer, and when implemented, the clustering method may include:

s1: calculating the distance between the current clustering center and the last clustering center; judging whether the distance is smaller than a first preset difference value or not;

s2: taking the last classification result as a final classification result and stopping clustering under the condition that the distance is smaller than a first preset difference value;

s3: and under the condition that the distance is greater than or equal to the first preset difference value, dividing each client into a cluster category corresponding to the current-time cluster center with the minimum distance to the client according to the distance between each client and each current-time cluster center, and taking the cluster category as the current-time classification result of the client.

In some embodiments, before obtaining the last cluster center, the method further comprises:

s1: judging whether the current round is the first round or not;

s2: taking the initial clustering center as the last clustering center under the condition that the current round is the first round;

s3: and under the condition that the current round is not the first round, determining to acquire the last clustering center.

Based on the embodiment, the problem that the clustering result is not ideal because the initial clustering center is randomly selected by the traditional K-means clustering algorithm is solved; in addition, the influence of the last clustering center is considered in the calculation process of the clustering center, so that the classification result is more accurate.

In some embodiments, after the obtaining of the classification result of the customer by performing a preset clustering process on the customer according to the initial clustering center and the value vector of the customer, the method further includes:

s1: determining the clients contained in different client type groups according to the classification result of the clients;

s2: configuring matched promotion data according to different client types;

s3: and pushing matched promotion data to the clients contained in the different client type groups.

Based on the embodiment, the clients can be accurately classified according to the data mining technology to obtain the classification results of different clients, matched promotion data are pushed to different client type groups according to the accurate classification results of the clients, a good pushing effect is obtained, the manpower service cost is reduced, and the bank profits are improved.

In a specific scenario example, referring to fig. 2, a bank customer may be classified by using the customer classification method provided in this specification. Before specific implementation, a current value characteristic index, a potential value characteristic index and a dependency characteristic index of a client are selected, wherein the current value characteristic index can be a target recent transaction time value, a target transaction frequency value and a target total transaction amount, the potential value characteristic index can be a quantized age value, a quantized gender value, an education background value and a quantized income level value, and the dependency characteristic index can be a psychological dependency characteristic and a behavior dependency characteristic; then, a value vector model is constructed according to the current value characteristics, the potential value characteristics and the dependency characteristics of the client and the corresponding calculation method, for example, the calculation method corresponding to the current value characteristics can be an RFM model; in specific implementation, the customers are classified according to the value vector model to obtain accurate classification results of the customers, wherein the classification algorithm can be an improved K-means algorithm, and the accurate classification results of the customers can be used for a bank to push matched popularization data for different customer type groups.

Although the present specification provides method steps as described in the examples or flowcharts, additional or fewer steps may be included based on conventional or non-inventive approaches. The order of steps recited in the embodiments is merely one manner of performing the steps in a multitude of orders and does not represent the only order of execution. When an apparatus or client product in practice executes, it may execute sequentially or in parallel (e.g., in a parallel processor or multithreaded processing environment, or even in a distributed data processing environment) according to the embodiments or methods shown in the figures. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, article, or apparatus that comprises the recited elements is not excluded. The terms first, second, etc. are used to denote names, but not any particular order.

Based on the above customer classification method, one or more embodiments of the present specification further provide a customer classification device. The apparatus may include systems (including distributed systems), software (applications), modules, components, servers, clients, etc. that use the methods described in the embodiments of the present specification in conjunction with any necessary apparatus to implement the hardware. Based on the same innovative conception, embodiments of the present specification provide an apparatus as described in the following embodiments. Because the implementation scheme of the apparatus for solving the problem is similar to that of the method, the specific apparatus implementation in the embodiment of the present description may refer to the implementation of the foregoing method, and repeated details are not described again. As used hereinafter, the term "unit" or "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.

Specifically, fig. 3 is a schematic block diagram of an embodiment of a customer sorting apparatus provided in this specification, and as shown in fig. 3, the customer sorting apparatus provided in this specification may include: the center generating module 301, the obtaining module 302, the constructing module 303 and the classifying module 304.

A center generation module 301, configured to determine a client type for a client; determining an initial clustering center according to the customer type;

an obtaining module 302, configured to obtain attribute data and behavior data of a client;

the building module 303 is configured to build a value vector of the customer by using the attribute data and the behavior data of the customer according to a preset processing rule;

and the classification module 304 is configured to perform preset clustering processing on the customer according to the initial clustering center and the value vector of the customer to obtain a classification result of the customer.

It should be noted that the above-mentioned description of the apparatus according to the method embodiment may also include other embodiments, and specific implementation manners may refer to the description of the related method embodiment, which is not described herein again.

The present specification also provides a server comprising a processor and a memory for storing processor-executable instructions, which when executed by the processor, implement any of the above method embodiments. For example, the instructions when executed by the processor implement steps comprising: determining a customer type for the customer; determining an initial clustering center according to the customer type; acquiring attribute data and behavior data of a client; constructing a value vector of the client by utilizing the attribute data and the behavior data of the client according to a preset processing rule; and according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer.

It should be noted that the above-mentioned server may also include other implementations according to the description of the method or apparatus embodiment. The specific implementation manner may refer to the description of the related method embodiment, and is not described in detail herein.

Specifically, fig. 4 is a block diagram of a hardware structure of an embodiment of a server provided in this specification. As shown in fig. 4, the server 10 may include one or more (only one shown) processors 100 (the processors 100 may include, but are not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 200 for storing data, and a transmission module 300 for communication functions. It will be understood by those skilled in the art that the structure shown in fig. 4 is only an illustration and is not intended to limit the structure of the electronic device. For example, the server 10 may also include more or fewer components than shown in FIG. 4, and may also include other processing hardware, such as a database or multi-level cache, a GPU, or have a different configuration than shown in FIG. 4, for example.

The memory 200 may be used to store software programs and modules of application software, such as program instructions/modules corresponding to the customer classification method in the embodiment of the present specification, and the processor 100 executes various functional applications and data processing by operating the software programs and modules stored in the memory 200. Memory 200 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 200 may further include memory located remotely from processor 100, which may be connected to a computer terminal over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission module 300 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the computer terminal. In one example, the transmission module 300 includes a Network adapter (NIC) that can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission module 300 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.

The present specification also provides a computer readable storage medium having stored thereon computer program instructions that, when executed, implement: determining a customer type for the customer; determining an initial clustering center according to the customer type; acquiring attribute data and behavior data of a client; constructing a value vector of the client by utilizing the attribute data and the behavior data of the client according to a preset processing rule; and according to the initial clustering center and the value vector of the customer, carrying out preset clustering processing on the customer to obtain a classification result of the customer.

The foregoing description of specific embodiments has been presented for purposes of illustration and description. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, 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 method or apparatus provided by the present specification and described in the foregoing embodiments may implement service logic through a computer program and record the service logic on a storage medium, where the storage medium may be read and executed by a computer, so as to implement the effect of the solution described in the embodiments of the present specification. The storage medium may include a physical device for storing information, and typically, the information is digitized and then stored using an electrical, magnetic, or optical media. The storage medium may include: devices that store information using electrical energy, such as various types of memory, e.g., RAM, ROM, etc.; devices that store information using magnetic energy such as hard disks, floppy disks, tapes, core memories, bubble memories, and usb disks; devices that store information optically, such as CDs or DVDs. Of course, there are other ways of storing media that can be read, such as quantum memory, graphene memory, and so forth.

The embodiment of the client classification method or apparatus provided in this specification may be implemented in a computer by a processor executing corresponding program instructions, for example, by using c + + language of a windows operating system to be implemented on a PC side, a linux system, or by using android and iOS system programming languages to be implemented on an intelligent terminal, or by using processing logic of a quantum computer.

It should be noted that descriptions of the apparatuses and devices described above according to the related method embodiments in the specification may also include other embodiments, and specific implementation manners may refer to descriptions of corresponding method embodiments, which are not described in detail herein.

The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on differences from other embodiments. In particular, for the hardware + program class embodiment, since it is substantially similar to the method embodiment, the description is simple, and reference may be made to part of the description of the method embodiment for relevant points.

For convenience of description, the above devices are described as being divided into various modules by functions, which are described separately. Of course, when implementing one or more of the present description, the functions of some modules may be implemented in one or more software and/or hardware, or the modules implementing the same functions may be implemented by a plurality of sub-modules or combinations of sub-units, etc.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus, devices according to embodiments of the invention. It will be understood that they may be implemented by computer program instructions which may be provided to a processor of a general purpose computer, special purpose computer, embedded processor or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

One skilled in the art will appreciate that one or more embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, one or more embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects.

While the specification has been described with examples, those skilled in the art will appreciate that there are numerous variations and permutations of the specification that do not depart from the spirit of the specification, and it is intended that the appended claims include such variations and modifications that do not depart from the spirit of the specification.

Claims

1. A customer categorization method, comprising:

acquiring attribute data and behavior data of a client;

2. The method of claim 1, wherein the customer type of the customer comprises: the client system comprises clients with high current value, high potential value and high dependency, clients with high current value, high potential value and low dependency, clients with high current value, low potential value and high dependency, clients with high current value, low potential value and low dependency, clients with low current value, high potential value and high dependency, clients with low current value, high potential value and low dependency, clients with low current value, low potential value and high dependency, and clients with low current value, low potential value and low dependency.

3. The method of claim 1, wherein constructing a value vector for a customer using the customer's attribute data and behavior data according to a predetermined processing rule comprises:

according to a preset processing rule, determining the current value characteristic, the potential value characteristic and the dependency characteristic of the client by using the attribute data and the behavior data of the client respectively;

and combining the current value characteristic, the potential value characteristic and the dependency characteristic of the client to obtain a value vector of the client.

4. The method of claim 3, wherein determining the current value characteristics of the customer using the customer's attribute data and behavior data according to preset processing rules comprises:

obtaining a latest transaction time equal partition, a transaction frequency equal partition and a total transaction amount equal partition according to latest transaction time data, transaction frequency data and total transaction amount data in the behavior data of the client;

5. The method of claim 3, wherein determining the potential value characteristics of the customer using the customer's attribute data and behavior data according to predetermined processing rules comprises:

and determining the potential value characteristics of the client according to the quantified age value, gender value, education background value and income level value.

6. The method of claim 1, wherein the obtaining the classification result of the customer by performing a predetermined clustering process on the customer according to the initial clustering center and the value vector of the customer comprises:

and according to the initial clustering center and the value vector of the client, carrying out preset clustering processing on the client by utilizing an improved K-means algorithm to obtain a classification result of the client.

7. The method of claim 6, wherein the obtaining the classification result of the customer by performing a predetermined clustering process on the customer using the modified K-means algorithm comprises:

acquiring a last clustering center and a last classification result;

8. The method of claim 1, wherein after obtaining the classification result of the customer by performing a predetermined clustering process on the customer based on the initial clustering center and the value vector of the customer, the method further comprises:

configuring matched promotion data according to different client types;

9. A customer sorting apparatus, comprising:

the acquisition module is used for acquiring attribute data and behavior data of a client;

10. A server comprising a processor and a memory for storing processor-executable instructions which, when executed by the processor, implement the steps of the method of any one of claims 1 to 8.

11. A computer-readable storage medium having stored thereon computer instructions which, when executed, implement the steps of the method of any one of claims 1 to 8.