Client type identification method and device

Abstract

The invention discloses a customer type identification method and device, and relates to the technical field of big data, and the method comprises the steps: obtaining customer information; respectively matching the positioning information of the mobile terminal and the registration place information of the electronic bank system of the customer with a place name database, and when the matching result is that the positioning information of the mobile terminal and the registration place information of the electronic bank system of the customer are matched with the place name information of a specified type, acquiring family information of the customer according to the unique identifier of the customer; analyzing family information of the customers in a pre-constructed customer relationship network by using a graph theory algorithm, and classifying the customers into the families to which the customers belong; analyzing the Enger coefficient of the family to which the customer belongs; and determining the customer type according to the Enger coefficient of the family to which the customer belongs, so that the accuracy and efficiency of determining the customer type can be improved.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to a customer type identification method and device. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] The rapid development of the Internet has led to various industries competing to launch their own clients, customize relevant marketing strategies in the clients according to the characteristics of their own industries and customers, and launch relevant marketing activities. However, since the download, registration, and login of the client are all self-service, it is impossible to identify whether the customer belongs to the target customer of the client. At the same time, since the client often launches indiscriminate marketing activities such as coupon collection when it goes online, the client's customer group is mixed with some non-target customers of marketing activities. Therefore, when performing data analysis during the production and operation process, it is impossible to judge whether the current client business situation truly reflects the actual needs of the target customers based on the background data, and it is impossible to further iterate and develop. The existing technology often fills in relevant information and adds corresponding tags when the customer logs in. Not only is the efficiency low, but also due to the uncertainty of crowd flow and tag timeliness, the accuracy of identifying customer types through tags is greatly reduced over time. The existing technology also uses expert rules to identify customer types, but because the rules are often limited by subjectivity, they can only identify target groups to a limited extent. As time goes by, it also takes manpower to update and maintain expert rules to maintain their effectiveness, and the accuracy and efficiency of customer type identification are low. Summary of the invention

[0004] An embodiment of the present invention provides a method for identifying a customer type, for improving the accuracy and efficiency of identifying a customer type, the method comprising:

[0005] Obtain customer information; customer information includes the location information of the mobile terminal, the customer's electronic banking system registration location information, and the customer's unique identification in the electronic banking system;

[0006] Matching the location information of the mobile terminal and the customer's electronic banking system registration location information with a place name database, where different types of place name information are recorded;

[0007] When the matching result is that the positioning information of the mobile terminal and the electronic banking system registration location information of the customer both match the designated type of place name information, the family information of the customer is obtained in the electronic banking system according to the unique identification of the customer;

[0008] By using a graph theory algorithm, in a pre-constructed customer relationship network, the customer's family information is analyzed and the customer is classified into the corresponding family; the customer relationship network is constructed based on the analysis of the family information of historical customers;

[0009] Analyze the Engel coefficient for the customer's family;

[0010] Determine the customer type based on the Engel coefficient of the family to which the customer belongs.

[0011] The embodiment of the present invention further provides a customer type identification device for improving the accuracy and efficiency of identifying customer types, the device comprising:

[0012] The customer information acquisition module is used to acquire customer information; the customer information includes the location information of the mobile terminal, the customer's electronic banking system registration location information, and the customer's unique identification in the electronic banking system;

[0013] A matching module, used to match the location information of the mobile terminal and the customer's electronic banking system registration location information with a place name database, wherein different types of place name information are recorded in the place name database;

[0014] A family information acquisition module, used to acquire the family information of the customer in the electronic banking system according to the unique identification of the customer when the matching result is that the positioning information of the mobile terminal and the registration place information of the customer's electronic banking system both match the place name information of the specified type;

[0015] A family analysis module is used to analyze the family information of customers in a pre-built customer relationship network using a graph theory algorithm and classify the customers into their respective families; the customer relationship network is constructed based on the analysis of family information of historical customers;

[0016] Engel coefficient analysis module, used to analyze the Engel coefficient of the customer's family;

[0017] The determination module is used to determine the customer type according to the Engel coefficient of the family to which the customer belongs.

[0018] An embodiment of the present invention further provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned customer type identification method when executing the computer program.

[0019] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned customer type identification method is implemented.

[0020] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned customer type identification method is implemented.

[0021] Compared with the prior art, the embodiment of the present invention obtains customer information; the customer information includes the positioning information of the mobile terminal, the customer's electronic banking system registration place information, and the customer's unique identification in the electronic banking system; the positioning information of the mobile terminal and the customer's electronic banking system registration place information are respectively matched with a place name database, and the place name database records different types of place name information; when the matching result is that the positioning information of the mobile terminal and the customer's electronic banking system registration place information are both matched with the specified type of place name information, the customer's family information is obtained in the electronic banking system according to the customer's unique identification; using a graph theory algorithm, in a pre-constructed customer relationship network, the customer's family information is analyzed and the customer is classified into a corresponding family; the customer relationship network is constructed based on the analysis of historical customer family information; the Engel coefficient of the customer's family is analyzed; and the customer type is determined based on the Engel coefficient of the customer's family, so as to improve the accuracy and efficiency of identifying the customer type. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0023] Figure 1 is a flow chart of a method for identifying a customer type in an embodiment of the present invention;

[0024] Figure 2 A flowchart of a specific example of a method for identifying a customer type in an embodiment of the present invention;

[0025] Figure 3 Schematic diagram of a customer type identification device in an embodiment of the present invention;

[0026] Figure 4 is a schematic diagram of a specific example of a customer type identification device in an embodiment of the present invention;

[0027] Figure 5 Schematic diagram of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0029] In view of the problem that the download, registration and login of the client are all done by the client themselves, it is impossible to identify whether the client is the target customer of the client. The embodiment of the present invention provides a method and device for identifying the type of the client who logs in to the client, so as to identify the target customer from the client's customer group mixed with some non-target customers of marketing activities. When performing data analysis during the production operation process, it is possible to judge whether the current client business situation truly reflects the actual needs of the target customer based on the background data, so as to further iterate the research and development of the client. Compared with the existing technology, the accuracy and efficiency of customer type identification are improved.

[0030] Specifically, the embodiment of the present invention uses farmer customers as target customers to illustrate the implementation of the customer type identification method. For example, a certain client is a mobile banking client, whose purpose is to achieve rural development through banking services, and the target customers are farmer customers. There are often some non-farmer customers in the customer group. When non-farmer customers download, log in and use the client targeted for farmer customers on the mobile client, its background operation data has no reference value for the production operation in the direction of rural development, and it is impossible to judge whether the current client business situation truly reflects the actual needs of the target customers, namely farmer customers, based on the background data, which further causes this type of client to be unable to iterate and update.

[0031] To solve the above problems, the present invention provides a customer type identification method, which can improve the accuracy of determining customer types and ensure the targeted analysis of client data. It can help enterprises identify customer types from massive customer data and select target customers from logged-in customers, so as to further conduct production operation analysis on target customers and ensure the steady iteration and upgrade of the client.

[0032] Figure 1 FIG. 1 is a flow chart of a method for identifying a customer type in an embodiment of the present invention. Figure 1 As shown, the customer type identification method includes:

[0033] Step 101, obtaining customer information; the customer information includes the location information of the mobile terminal, the customer's electronic banking system registration location information, and the customer's unique identification in the electronic banking system;

[0034] Step 102, matching the location information of the mobile terminal and the customer's electronic banking system registration location information with a place name database, wherein different types of place name information are recorded in the place name database;

[0035] Step 103, when the matching result is that the positioning information of the mobile terminal and the electronic banking system registration location information of the customer both match the designated type of place name information, the family information of the customer is obtained in the electronic banking system according to the unique identification of the customer;

[0036] Step 104, using a graph theory algorithm, analyzing the customer's family information in a pre-constructed customer relationship network, and classifying the customer into the family to which he belongs; the customer relationship network is constructed based on the analysis of the family information of historical customers;

[0037] Step 105, analyzing the Engel coefficient for the household to which the customer belongs;

[0038] Step 106, determining the customer type according to the Engel coefficient of the family to which the customer belongs.

[0039] pass Figure 1 The customer type identification method can improve the accuracy and efficiency of determining the customer type and ensure the pertinence of the client data analysis. Specifically, it can help banks identify customer types from massive customer data and select target customers, such as farmer customers, from the logged-in customers, so as to further conduct production operation analysis for the target customers and ensure the steady iteration and upgrade of the client. In one embodiment, the customer uses a mobile terminal to log in to the electronic banking system, which can be the customer logging in to the bank app on the mobile terminal.

[0040] In one embodiment, in step 101, when a customer logs into the electronic banking system using a mobile terminal, a request for obtaining customer information can be sent to the mobile terminal, and after authorization by the customer, the customer's mobile phone location information, the customer's electronic banking system registration location information, and the customer's unique identification (ID) in the electronic banking system can be obtained.

[0041] In one embodiment, in step 102, the customer's location information and the customer's bank App registration location information can be matched with the place name database, which records different types of place name information. For example, for a bank App whose target customers are farmers, the customer's location information and the customer's bank App registration location information can be matched with the village list, and both must belong to the content in the village list, that is, if the conditions are met, the subsequent step 103 is further judged; if either does not belong, it is considered that the logged-in customer is not a farmer customer, and the result is output without the need for subsequent steps.

[0042] In one embodiment, when step 102 determines that the customer has the possibility of being a target customer, step 103 can be executed to obtain the customer's family information in the electronic banking system based on the customer's unique identifier. The customer information can be obtained on a per-customer basis, that is, on a per-customer basis, that is, on a unique customer ID basis, especially the home address, home phone number, family members, and other information that can reflect the customer's social and family relationships, to build a customer relationship information database, and to maintain the classification of these information types.

[0043] In one embodiment, before executing step 104, i.e. analyzing the customer's family information, the following may also be included: obtaining historical customer information, the historical customer information including the historical customer's unique customer identification in the electronic banking system; obtaining the historical customer's family information in the electronic banking system according to the historical customer's unique customer identification; the historical customer's family information including one or any combination of family address information, family phone information, and family member information; constructing an initial customer relationship network according to the historical customer information and the historical customer's family information; and using a graph theory algorithm to perform community structure analysis on the initial customer relationship network to obtain a customer relationship network. Pre-constructing a customer relationship network using massive historical customer information data is beneficial to subsequent type recognition of logged-in customers and improves recognition accuracy.

[0044] In this embodiment, obtaining historical customer information may include: merging multiple account data of historical customers based on the unique customer identification of the historical customers in the electronic banking system. Since each customer has a unique customer ID in the electronic banking system, which represents the unique identification between different customers, the information data of the same customer may be merged based on the customer ID, including the case where multiple accounts, cards, mobile phone numbers, etc. belong to the same customer. In this case, the merged data set will be classified by customer to avoid information duplication and make subsequent analysis more accurate.

[0045] In this embodiment, since fields such as home address, home phone number, family members, etc. in the customer information are usually filled in by the customers themselves, there may be problems such as invalidity and uselessness. Therefore, the data set needs to be cleaned and pre-processed, including removing useless fields in the customer information, processing missing and outliers, etc., and further automatically verifying the information through semantic recognition and crawler software to eliminate situations where the customer's random filling in information, such as the corresponding street does not exist in the city, the house number is too long and invalid, etc., to ensure that all information in the data set classified by the customer dimension is true and valid.

[0046] In this embodiment, the method for obtaining the family information of the customer is the same as that of obtaining the family information of the historical customer in the electronic banking system according to the unique customer identifier of the historical customer. The family information of the historical customer can be obtained on a unit basis, that is, on a unit basis, based on the unique customer ID, especially the home address, home phone number, family members and other information that can reflect the social and family relationships of the customer, to build a historical customer relationship information database and maintain the classification of these information types.

[0047] In one embodiment, since the family is required for subsequent customer type identification, the initial customer relationship network can be selected by analyzing the information similarity to improve the practicality of the initial customer relationship network. The initial customer relationship network can be constructed based on historical customer information and historical customer family information, which can include: determining the similarity of the same type of information in the historical customer family information; using the historical customers corresponding to the family information with a similarity exceeding a preset threshold as nodes of the initial customer relationship network; based on the historical customer family information, using the family relationship between historical customers corresponding to the information with a similarity exceeding a preset threshold as the edge of the node in the initial customer relationship network.

[0048] In the customer relationship database, the information of each customer under the same information type (such as home address, home phone number, family members, etc.) is traversed to identify whether there is common information or information with a similarity exceeding a preset threshold (which can be flexibly set according to actual business needs). If so, the relationship between these customers is established into an initial network. The customer relationship in the electronic banking system can also be converted into a graph structure representation. In this graph, each customer is represented as a node, and the relationship between nodes can be constructed through common home address, home phone number, family members, etc. information, as a node (customer) connected to each other in the network.

[0049] In one embodiment, in order to divide customers into reasonable family groups, improve the accuracy of family member classification, and improve the accuracy of customer type identification, a graph theory algorithm is used to perform community structure analysis on an initial customer relationship network to obtain a customer relationship network, which may include: determining the degree of nodes in the initial customer relationship network; taking each node as a community to establish an initial community set; repeating the following steps until all edges in the customer relationship network are removed or the maximum value of the modularity of all communities no longer increases: determining the betweenness of edges in the customer relationship network; removing edges with the largest betweenness; determining the modularity of each community; merging communities with the largest modularity into new communities to establish a new community set.

[0050] For each node, calculate the number of edges connected to it. In graph theory algorithms, the degree of nodes is used to sort and divide communities. Calculate the betweenness of all edges to indicate the importance of the edge on the network. Betweenness usually indicates the degree to which a node or edge connects other nodes in the network. Because the betweenness algorithm can easily lead to problems of computational complexity and time complexity, some algorithms based on approximate calculations, such as the Brandes algorithm, need to be adopted. After the edge is removed, the connectivity of the network graph of the customer relationship network will be destroyed, and the graph will be divided into two or more subgraphs. In order to improve the efficiency of the algorithm and reduce errors, the weight of the edge needs to be adjusted in reverse during this process. Calculating the modularity of each community can divide customers into reasonable family groups. After no edges can be removed or the maximum modularity value no longer increases, the final community division result, that is, the latest community set, is obtained.

[0051] In this embodiment, in order to more conveniently find customers who belong to the same family as the customer in the future to calculate the family Engel coefficient, the communities with the largest modularity are merged into a new community, and a new community set is established, which may include: treating the historical customers corresponding to the nodes divided into the same community in the new community set as a family; setting the same family ID for the historical customers belonging to the same family. For customers from different families, their family IDs are different. For customers belonging to the same family, a family relationship is established, including information such as family members. Customers belonging to the same family can be identified by the family ID.

[0052] In this embodiment, in order to analyze the Engel coefficient, the family ID can be used to obtain the consumption records of family members and then calculate it. For the family to which the customer belongs, analyzing the Engel coefficient can include: obtaining the customer's unique identification in the electronic banking system for the customer who has the same family ID as the customer; obtaining the customer's consumption record in the electronic banking system based on the customer's unique identification; obtaining the merchant type corresponding to the consumption record in the electronic banking system; adding food tags to the consumption records of merchants whose merchant types belong to food merchants; determining the proportion of the total consumption of consumption records with food tags to the total consumption of all consumption records within a preset time period; and determining the Engel coefficient of the family to which the customer belongs based on the proportion.

[0053] The characteristics of the Engel coefficient of different types of groups are obtained. For example, the Engel coefficient of a general rural family is higher than that of an urban family. There is an approximate range for different types. The customer type is determined by analyzing the Engel coefficient in combination with the matching result of step 102. The customer ID of each member is obtained through the family ID, so as to obtain all the historical consumption records of each member in the bank database. For each consumption record, the merchant type corresponding to the consumption record in the system is obtained, and the merchant type is automatically determined whether it belongs to a food merchant through semantic recognition or calling databases such as "Tianyancha" and "Qichacha". For consumption records belonging to food merchants, food tags are added to the consumption records. After the above steps traverse all the historical consumption records of the family, the consumption records containing food tags are separately classified. Calculate the total consumption of food classification and calculate the total historical consumption of the family. Calculate the proportion of the total consumption of food classification to the total historical consumption of the family, and return the proportion result as the Engel coefficient of the family to which the current logged-in customer belongs, and preliminarily determine whether the logged-in customer belongs to the target customer through the Engel coefficient. Calculation formula: Engel coefficient = (family food expenditure / family total expenditure) × 100%; where, family food expenditure = family member A's food expenditure + family member B's food expenditure + ... + family member N's ​​food expenditure; family total expenditure = family member A's total expenditure + family member B's total expenditure + ... + family member N's ​​total expenditure. For example, the Engel coefficient of a general rural family is between 40% and 50%. For a client whose target customers are farmers, if the customer's Engel coefficient is close to or between 40% and 50% (it can also be flexibly configured according to actual business needs), then the customer is considered to meet the conditions and is the target customer of the client.

[0054] The implementation of the customer type identification method is further introduced below with a specific embodiment in which the target customers are farmers:

[0055] Figure 2 FIG. 1 is a flowchart of a specific example of a method for identifying a customer type in an embodiment of the present invention. Figure 2 As shown, after the customer logs in to the mobile terminal, the embodiment of the present invention determines the customer's location and calculates the Engel system to determine whether the customer is a farmer customer, as follows:

[0056] 1. Obtain customer information

[0057] When a customer logs in using a mobile terminal, we obtain the customer's login information, which includes three aspects: the customer's mobile phone location information, the customer's electronic banking registration location, and the customer's unique identification in the electronic banking system, such as the customer number.

[0058] After obtaining the information, two aspects of judgment are performed separately and simultaneously. One is step 2, and the other is step 3 and step 4. If both aspects are met, the customer is determined to be a farmer customer. If only one of them is met, it is determined that there is a possibility that the customer is a farmer customer. If none of them are met, it is determined that the customer is not a farmer customer.

[0059] 2. Determine the customer's location

[0060] The customer's mobile phone location information and electronic banking registration location information are matched with the village list respectively. Both must belong to the content on the list. If the conditions are met, proceed to the subsequent steps for further judgment; if either does not belong, the customer is considered not to be a farmer customer, and the result is output without the need for subsequent steps.

[0061] 3. Classification of family members

[0062] Since step 4 will use the Engel coefficient for judgment, and the Engel coefficient is calculated based on the household, this step is used to classify different customers into households by obtaining the customer information such as home address, home phone number, family members, etc. in the electronic banking system, as well as the common information associated between different accounts:

[0063] 3.1. Establish customer data set

[0064] In the electronic banking system, each customer has a unique customer ID, which is a unique identifier between different customers. The information data of the same customer is merged based on the customer ID, including the situation where multiple accounts, cards, mobile phone numbers, etc. belong to the same customer. In this case, the merged data set will be classified by customer, and there will be no information overlap.

[0065] 3.2. Data cleaning and preprocessing

[0066] Since fields such as home address, home phone number, and family members in customer information are usually filled in by the customers themselves, there may be problems such as invalidity or uselessness. Therefore, the data set needs to be cleaned and preprocessed, including removing useless fields in customer information, processing missing and outliers, and further automatically verifying the information through semantic recognition and crawler software to eliminate customers' random information such as the non-existence of corresponding streets in the city and the invalid house number that is too long, so as to ensure that all information in the data set classified by customer dimensions is true and valid.

[0067] 3.3. Acquisition of customer relationship information

[0068] Taking each customer as a unit, that is, taking the unique customer ID as a unit, obtain customer information, especially the home address, home phone number, family members and other information that can reflect the customer's social and family relationships, build a customer relationship information database, and maintain the classification of these information types.

[0069] 3.4.Build customer relationship network

[0070] In the customer relationship database, the information of each customer under the same information type (such as home address, home phone number, family members, etc.) is traversed to identify whether there is common information or information with a similarity greater than α (which can be flexibly set according to actual business needs). If so, the relationship between these customers is established into a network, which can be constructed in a way of seeking common ground while reserving differences, that is, if the single information type is the same, it is determined that a relationship network can be constructed (such as the case where the home address is the same but the home phone number and family members are different), and the customer relationship in the electronic banking system is converted into a graph structure representation. In this graph, each customer is represented as a node, and the relationship between nodes can be constructed through common home address, home phone number, family members and other information, as a node (customer) connected to each other in the network.

[0071] 3.5. Graph Theory Algorithm Analysis

[0072] Utilize graph theory algorithms, such as the community discovery method based on the GN algorithm in graph theory algorithms, to analyze and classify customer relationship networks:

[0073] 3.5.1. Calculating the degree of a node

[0074] For each node, calculate its degree, which is the number of edges connected to it. In the GN algorithm, the degree of the node is used to sort and divide the community.

[0075] 3.5.2. Initialize the community

[0076] All nodes are considered as separate communities, forming the initial community set.

[0077] 3.5.3. Calculating edge betweenness

[0078] Calculate the betweenness of all edges to indicate the importance of the edge in the graph. Betweenness usually indicates the degree to which a node or edge connects other nodes in the graph. Because betweenness algorithms can easily lead to problems with computational complexity and time complexity, it is necessary to use some algorithms based on approximate calculations, such as the Brandes algorithm.

[0079] 3.5.4. Remove edges

[0080] Remove edges in descending order of betweenness, one edge at a time. After the edge is removed, the connectivity of the graph will be destroyed, and the graph will be divided into two or more subgraphs. In order to improve the efficiency of the algorithm and reduce errors, the weights of the edges need to be adjusted in reverse during this process.

[0081] 3.5.5. Calculating modularity

[0082] Calculate the modularity of each community. Modularity indicates the closeness between nodes in the community. The higher the modularity, the closer the connection within the community. Modularity can be used to evaluate the quality of community division, and the ultimate goal is to divide customers into reasonable family groups.

[0083] 3.5.6. Merge communities

[0084] Merge the two communities with the largest modularity into a new community to form a new community set. Repeat the above steps 3.5.3-3.5.6 until no edges can be removed or the maximum modularity no longer increases.

[0085] 3.6. Create a Family ID

[0086] The final community division result of step 3.5 is used to classify customer family information. For customers belonging to the same family, they are set to the same family ID. For customers from different families, their family IDs are different.

[0087] 3.7. Establishing family relationships

[0088] For customers belonging to the same family, a family relationship is established, including information such as family members. Customers belonging to the same family can be identified by the family ID.

[0089] 4. Determine the Engel coefficient

[0090] The Engel coefficient is an indicator used to measure consumption structure, which indicates the percentage of a family's food expenditure in total expenditure. According to existing data, the Engel coefficients of urban and rural residents are generally as follows:

[0091] Engel coefficient of rural residents: 40% to 50%

[0092] Engel coefficient of urban residents: 30% to 40%

[0093] 4.1. Obtain consumption records

[0094] The customer number of each member is obtained through the family classified in step 3, so as to obtain all historical consumption records of each member in the bank database.

[0095] 4.2. Get consumption type

[0096] For each consumption record, the merchant type corresponding to the consumption record in the system is obtained, and through semantic recognition or calling databases such as "Tianyancha" and "Qichacha" to automatically determine whether the merchant type belongs to a food merchant.

[0097] 4.3. Add food label

[0098] For consumption records belonging to food merchants, add food tags to the consumption records.

[0099] 4.4. Classified consumption records

[0100] After the above steps traverse all historical consumption records of the family, the consumption records containing food tags are classified separately.

[0101] 4.5. Calculate the consumption amount

[0102] Calculate the total consumption of food categories and the total historical consumption of the household.

[0103] 4.6. Calculation of Engel coefficient

[0104] Calculate the proportion of the total consumption of food categories to the total historical consumption of the family, and return the proportion result as the Engel coefficient of the family to which the current logged-in customer belongs. The Engel coefficient is used to preliminarily determine whether the logged-in customer is a farmer.

[0105] Calculation formula:

[0106] Engel coefficient = household food expenditure / total household expenditure × 100

[0107] Family food expenditure = Family member A's food expenditure + Family member B's food expenditure + ... + Family member N's ​​food expenditure

[0108] Food expenditure of family member X = consumption record 1 of family member X (with food mark) + consumption record 2 of family member X (with food mark) + ... + consumption record N of family member X (with food mark)

[0109] Total household expenditure = total expenditure of family member A + total expenditure of family member B + ... + total expenditure of family member N

[0110] Total expenditure of family member X = consumption record 1 of family member X + consumption record 2 of family member X + ... + consumption record N of family member X

[0111] If the customer's Engel coefficient is close to or between 40% and 50% (it can also be flexibly configured according to actual business needs), the customer is considered to meet the conditions.

[0112] 5. Comprehensive judgment

[0113] Based on the above two aspects, one is step 2, and the other is step 3 and step 4. If both aspects are met at the same time, the customer is deemed to be a farmer customer. If only one of them is met, it is determined that there is a possibility that the customer is a farmer customer. If none of them are met, it is determined that the customer is not a farmer customer.

[0114] The present invention also provides a client type identification device, as described in the following embodiments. Since the principle of the device to solve the problem is similar to that of the client type identification method, the implementation of the device can refer to the implementation of the client type identification method, and the repeated parts will not be repeated.

[0115] Figure 3 Schematic diagram of a customer type identification device in an embodiment of the present invention, comprising:

[0116] The customer information acquisition module 301 is used to acquire customer information; the customer information includes the location information of the mobile terminal, the customer's electronic banking system registration location information, and the customer's unique identification in the electronic banking system;

[0117] A matching module 302 is used to match the location information of the mobile terminal and the customer's electronic banking system registration location information with a place name database, where different types of place name information are recorded;

[0118] The family information acquisition module 303 is used to acquire the family information of the customer in the electronic banking system according to the unique identification of the customer when the matching result is that the positioning information of the mobile terminal and the registration place information of the customer's electronic banking system both match the place name information of the specified type;

[0119] The family analysis module 304 is used to analyze the family information of the customer in a pre-constructed customer relationship network by using a graph theory algorithm, and classify the customer into the family to which he belongs; the customer relationship network is constructed based on the analysis of the family information of historical customers;

[0120] The Engel coefficient analysis module 305 is used to analyze the Engel coefficient of the customer's family;

[0121] The determination module 306 is used to determine the customer type according to the Engel coefficient of the family to which the customer belongs.

[0122] Figure 4 FIG. 1 is a schematic diagram of a specific example of a customer type identification device in an embodiment of the present invention. Figure 4 As shown, in one embodiment, the customer type identification device further includes: a customer relationship network construction module 401, which is used to:

[0123] Obtaining historical customer information, including the unique identification of historical customers in the electronic banking system;

[0124] Acquiring family information of historical customers in the electronic banking system according to the unique identification of the historical customers; the family information of the historical customers includes one or any combination of family address information, family phone information, and family member information;

[0125] Build the initial customer relationship network based on historical customer information and historical customer family information;

[0126] Using graph theory algorithms, the community structure of the initial customer relationship network is analyzed to obtain the customer relationship network.

[0127] In one embodiment, the customer relationship network building module 401 is specifically used to:

[0128] The multiple account data of historical customers are merged based on their unique customer identification in the electronic banking system.

[0129] In one embodiment, the customer relationship network building module 401 is specifically used to:

[0130] Determine the similarity of the same type of information in the family information of historical customers;

[0131] The historical customers corresponding to the household information whose similarity exceeds a preset threshold are used as nodes of the initial customer relationship network;

[0132] According to the family information of historical customers, the family relationships between historical customers corresponding to the information with similarity exceeding a preset threshold are used as the edges of the nodes in the initial customer relationship network.

[0133] In one embodiment, the customer relationship network building module 401 is specifically used to:

[0134] Determine the degree of nodes in the initial customer relationship network;

[0135] Treat each node as a community and establish an initial community set;

[0136] Repeat the following steps until all edges in the customer relationship network are removed or the maximum value of the modularity of all communities no longer increases:

[0137] Determine the betweenness of edges in the customer relationship network;

[0138] Remove the edge with the largest betweenness;

[0139] Determine the modularity of each community;

[0140] Merge the communities with the largest modularity into new communities and establish new community sets.

[0141] In one embodiment, the customer relationship network building module 401 is specifically used to:

[0142] The historical customers corresponding to the nodes divided into the same community in the new community set are regarded as a family;

[0143] Set the same family ID for historical customers belonging to the same family.

[0144] In one embodiment, the family ID setting module 402 is specifically used to:

[0145] Obtain the unique identification of the customer in the electronic banking system that is the same as the customer's family ID;

[0146] Obtain the customer's consumption record in the electronic banking system based on the customer's unique identification;

[0147] Obtain the merchant type corresponding to the consumption record in the electronic banking system;

[0148] Add food tags to consumption records of food merchants;

[0149] Determine the proportion of the total consumption of consumption records with food marks to the total consumption of all consumption records within a preset period of time;

[0150] Based on the proportion, determine the Engel coefficient of the customer's household.

[0151] Compared with the prior art, the embodiment of the present invention obtains customer information; the customer information includes the positioning information of the mobile terminal, the customer's electronic banking system registration place information, and the customer's unique identification in the electronic banking system; the positioning information of the mobile terminal and the customer's electronic banking system registration place information are matched with the place name database respectively, and the place name database records different types of place name information; when the matching result is that the positioning information of the mobile terminal and the customer's electronic banking system registration place information are matched with the place name information of the specified type, the family information of the customer is obtained in the electronic banking system according to the customer's unique identification; the family information of the customer is analyzed in the pre-constructed customer relationship network by using the graph theory algorithm, and the customer is classified into the family to which he belongs; the customer relationship network is constructed according to the family information analysis of historical customers; the Engel coefficient of the customer's family is analyzed; the customer type is determined according to the Engel coefficient of the customer's family, which can improve the accuracy of determining the customer type, ensure the pertinence of the client data analysis, and ensure the steady iterative upgrade of the client. Specifically, it can help banks screen out target customers such as farmer customers from a large number of customers, ensure the pertinence of rural mobile terminal data analysis, and further develop steadily.

[0152] An embodiment of the present invention further provides a computer device, Figure 5 This is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 500 includes a memory 510, a processor 520, and a computer program 530 stored in the memory 510 and executable on the processor 520. When the processor 520 executes the computer program 530, the above-mentioned customer type identification method is implemented.

[0153] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned customer type identification method is implemented.

[0154] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the above-mentioned customer type identification method is implemented.

[0155] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0156] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0157] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0159] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for identifying customer types, characterized in that: include: Obtain customer information; customer information includes the location information of the mobile terminal, the customer's electronic banking system registration location information, and the customer's unique identification in the electronic banking system; Matching the location information of the mobile terminal and the customer's electronic banking system registration location information with a place name database, where different types of place name information are recorded; When the matching result is that the positioning information of the mobile terminal and the electronic banking system registration location information of the customer both match the designated type of place name information, the family information of the customer is obtained in the electronic banking system according to the unique identification of the customer; By using a graph theory algorithm, in a pre-constructed customer relationship network, the customer's family information is analyzed and the customer is classified into the corresponding family; the customer relationship network is constructed based on the analysis of the family information of historical customers; Analyze the Engel coefficient for the customer's family; Determine the customer type based on the Engel coefficient of the family to which the customer belongs.

2. The method according to claim 1, characterized in that Before analyzing the client's home information, it also includes: Obtaining historical customer information, including the unique identification of historical customers in the electronic banking system; Acquiring family information of historical customers in the electronic banking system according to the unique identification of the historical customers; the family information of the historical customers includes one or any combination of family address information, family phone information, and family member information; Build the initial customer relationship network based on historical customer information and historical customer family information; Using graph theory algorithms, the community structure of the initial customer relationship network is analyzed to obtain the customer relationship network.

3. The method according to claim 2, characterized in that Get historical customer information, including: The multiple account data of historical customers are merged based on their unique customer identification in the electronic banking system.

4. The method according to claim 2, characterized in that Build an initial customer relationship network based on historical customer information and historical customer family information, including: Determine the similarity of the same type of information in the family information of historical customers; The historical customers corresponding to the household information whose similarity exceeds a preset threshold are used as nodes of the initial customer relationship network; According to the family information of historical customers, the family relationships between historical customers corresponding to the information with similarity exceeding a preset threshold are used as the edges of the nodes in the initial customer relationship network.

5. The method according to claim 2, characterized in that Using graph theory algorithms, we conduct community structure analysis on the initial customer relationship network and obtain the customer relationship network, including: Determine the degree of nodes in the initial customer relationship network; Treat each node as a community and establish an initial community set; Repeat the following steps until all edges in the customer relationship network are removed or the maximum value of the modularity of all communities no longer increases: Determine the betweenness of edges in the customer relationship network; Remove the edge with the largest betweenness; Determine the modularity of each community; Merge the communities with the largest modularity into new communities and establish new community sets.

6. The method according to claim 5, characterized in that Merge the communities with the largest modularity into new communities and establish a new community set, including: The historical customers corresponding to the nodes divided into the same community in the new community set are regarded as a family; Set the same family ID for historical customers belonging to the same family.

7. The method according to claim 6, characterized in that Analyze the Engel coefficient for the customer's family, including: Obtain the unique identification of the customer in the electronic banking system that is the same as the customer's family ID; According to the unique identification of the customer, obtain the consumption record of the customer in the electronic banking system; Obtain the merchant type corresponding to the consumption record in the electronic banking system; Add food tags to consumption records of food merchants; Determine the proportion of the total consumption of consumption records with food marks to the total consumption of all consumption records within a preset period of time; Based on the proportion, determine the Engel coefficient of the customer's household.

8. A customer type identification device, characterized in that: include: The customer information acquisition module is used to acquire customer information; the customer information includes the location information of the mobile terminal, the customer's electronic banking system registration location information, and the customer's unique identification in the electronic banking system; A matching module, used to match the location information of the mobile terminal and the customer's electronic banking system registration location information with a place name database, wherein different types of place name information are recorded in the place name database; A family information acquisition module, used to acquire the family information of the customer in the electronic banking system according to the unique identification of the customer when the matching result is that the positioning information of the mobile terminal and the registration place information of the customer's electronic banking system both match the place name information of the specified type; A family analysis module is used to analyze the family information of customers in a pre-built customer relationship network using a graph theory algorithm and classify the customers into their respective families; the customer relationship network is constructed based on the analysis of family information of historical customers; Engel coefficient analysis module, used to analyze the Engel coefficient of the customer's family; The determination module is used to determine the customer type according to the Engel coefficient of the family to which the customer belongs.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images