Loan product recommendation method and device, electronic equipment and storage medium

Abstract

The invention discloses a loan product recommendation method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining customer information in response to a customer loan demand; and according to the customer information, obtaining a loan product matched with the customer based on a complex network community discovery model, the complex network community discovery model being used for discovering the similarity between the customers. According to the invention, different loan products can be recommended for different customer groups. According to the method and the device, more personalized and accurate product recommendation service can be realized.

Description

Technical Field

[0001] The present application relates to the technical field of loan product recommendation, and in particular to a loan product recommendation method, device, electronic device, and storage medium. Background Art

[0002] With the rapid development of Internet finance, the types of loan products are increasing. How to provide customers with more personalized and accurate product recommendations has become an important issue that needs to be solved in the field of Internet finance.

[0003] Most of the loan product recommendation methods in related technologies are based on simple customer information, loan history data, etc. However, this method has the following disadvantages:

[0004] a. The recommendation accuracy is not high. Common loan product recommendation methods are usually based on some simple rules and cannot take into account the multiple needs of customers and the dynamic changes in their needs.

[0005] b. Failure to consider differences among customer groups. Common loan product recommendation methods are mainly based on personal historical data, without considering the relationships and differences between customer groups. Summary of the invention

[0006] The embodiments of the present application provide a loan product recommendation method, device, electronic device, and storage medium to recommend different loan products to different customer groups, thereby achieving more personalized and accurate product recommendation services.

[0007] The present application embodiment adopts the following technical solutions:

[0008] In a first aspect, an embodiment of the present application provides a loan product recommendation method, wherein the recommendation method comprises:

[0009] Respond to customer loan requests and obtain customer information;

[0010] According to the customer information, a loan product matching the customer is obtained based on a complex network community discovery model, and the complex network community discovery model is used to discover similarities between customers.

[0011] In some embodiments, obtaining a loan product matching the customer based on the customer information by using a complex network community discovery model includes:

[0012] According to the customer information, the customers are regarded as nodes in a network and the associations between customers are regarded as edges in the network;

[0013] Divide nodes with similar characteristics in the network into different communities;

[0014] The complex network community discovery model is used to obtain similarities between customers in different communities, which is used to match loan products with customers.

[0015] In some embodiments, obtaining a loan product matching the customer based on the customer information by using a complex network community discovery model further includes:

[0016] Collect customer information in different communities and the types of loan products that customers have historically selected, and determine the loan products to be recommended;

[0017] Based on the proposed recommended loan product, a loan product matching the customer is obtained.

[0018] In some embodiments, in response to the customer's loan demand, obtaining customer information includes:

[0019] In response to customer loan demands, the Pearson correlation coefficient is used to calculate the correlation between customers and construct a customer relationship network.

[0020] In some embodiments, the step of obtaining customer information in response to a customer's loan demand further includes:

[0021] The Louvain algorithm is used to divide the customer relationship network into communities, thereby obtaining a modular structure of the customer relationship network.

[0022] In some embodiments, the step of obtaining customer information in response to a customer's loan demand further includes:

[0023] A characteristic analysis is performed on each community attribute, and the community attributes are compared, wherein the community attributes include at least one of the following: global efficiency, local efficiency, clustering coefficient, average shortest path length, and node degree.

[0024] In some embodiments, the method further comprises:

[0025] Pre-acquire personal information indicators, financial status indicators, credit record indicators, behavioral characteristic indicators, personal preference indicators, and loan history indicators;

[0026] in,

[0027] The personal information indicators are used to reflect the customer's personal situation and stability.

[0028] The financial status indicators are used to reflect the financial status and debt repayment ability of the customer.

[0029] The credit record indicator is used to reflect the customer's credit record and credit risk.

[0030] The behavioral characteristic indicators are used to reflect the behavioral characteristics and lifestyle of customers.

[0031] The personal preference index is used to reflect the personal preferences and needs of customers.

[0032] The loan history indicator is used to reflect the customer's historical loan situation.

[0033] In a second aspect, an embodiment of the present application further provides a loan product recommendation device, wherein the device comprises:

[0034] A response module, used to respond to customer loan demands and obtain customer information;

[0035] The matching recommendation module is used to obtain loan products that match the customer based on the customer information through a complex network community discovery method, and the complex network community discovery model is used to discover similarities between customers.

[0036] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to perform the above method.

[0037] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes the above method.

[0038] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects: in response to the customer's loan demand, obtain customer information, and obtain loan products matching the customer based on the customer information through a complex network community discovery model. Through the above method, the customer's needs and preferences can be better understood while reducing the complexity of the algorithm, thereby providing more accurate recommendation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0040] Figure 1 This is a flowchart of the loan product recommendation method in the embodiment of this application;

[0041] Figure 2 This is a schematic diagram of the structure of the loan product recommendation device in the embodiment of this application;

[0042] Figure 3 This is a schematic diagram of the structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0044] There are some product recommendation algorithms based on matrix decomposition, which regard customers and products as rows and columns in a matrix, and then use matrix decomposition to find the similarities between customers and products. The advantage of this method is that it is simple to implement, but its disadvantage is that there is a cold start problem. For example, when the recommendation system is just started, the system lacks customer historical behavior data, making it difficult to accurately predict customer interest in products. In addition, this method may also have an overfitting problem. When training the model, it is easy for the model to be too complex, thus overfitting the training data, and the generalization ability of the model decreases, resulting in deviations in the recommendation results. There are also some recommendation algorithms based on graph embedding, which regard customers and products as nodes in the graph, and then map them to a low-dimensional space through a graph embedding algorithm to find the similarities between them. The advantage of this method is that it can handle data sparsity and cold start problems, but its disadvantage is that the computational complexity is high.

[0045] In summary, the recommendation algorithms of related technologies mainly rely on traditional data mining and machine learning algorithms, and the recommendation results are often too general and cannot meet the personalized needs of customers. Secondly, the structure of complex network communities is complex and diverse. If it is difficult to accurately identify and analyze these communities, it will be impossible to provide customers with accurate loan product recommendations.

[0046] In response to these problems, the loan product recommendation method in the embodiment of the present application can more accurately identify and analyze the community where the customer is located by utilizing the characteristics and structural information of the complex network community, thereby providing customers with more personalized and accurate loan product recommendations. Compared with traditional data mining and machine learning algorithms, the method in the embodiment of the present application can better understand the needs and preferences of customers while reducing the complexity of the algorithm, thereby providing more accurate recommendation results.

[0047] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0048] This application embodiment provides a loan product recommendation method, such as Figure 1 As shown, a schematic diagram of the loan product recommendation method in an embodiment of the present application is provided, and the method at least includes the following steps S110 to S120:

[0049] Step S110, in response to the customer's loan demand, obtaining customer information.

[0050] Obtain user information based on customer loan needs. User information includes but is not limited to statistics and analysis results of customer characteristics from various angles to meet the needs of multi-dimensional customer information analysis and personalized recommendations. Specifically, it mainly includes the following information: name, age, gender, marital status, education level, occupation, income, expenditure, deposits, liabilities, credit rating, consumption habits, investment preferences, historical loan information, etc. In addition, it is necessary to pre-process the customer data in the customer loan demand to obtain customer information.

[0051] It should be noted that "customer loan demand" can be based on an Internet request or a local request.

[0052] Step S120: According to the customer information, a loan product matching the customer is obtained by using a complex network community discovery model, wherein the complex network community discovery model is used to discover similarities between customers.

[0053] Complex network community discovery technology is widely used in recommendation systems. Complex network community discovery is an analysis method based on network structure, which can divide nodes with similar characteristics in the network into different communities. The present invention regards customers as nodes in the network, and the associations between customers as edges in the network, and then uses the community discovery algorithm to discover the similarities between customers, so as to recommend the most suitable products to customers.

[0054] Based on the acquired customer information, the complex network community discovery model is used to match the loan products. The complex network community discovery model can find matching loan products by discovering the similarities between customers. When a customer needs a loan, the community to which the customer belongs is determined based on the customer's customer information, and then the theme loan products in the community are recommended.

[0055] By abstracting the relationship between customer information into a complex network model and analyzing the community structure and characteristics in the network, a comprehensive analysis and mining of customer groups is achieved.

[0056] Through the above method, according to the customer information, a loan product matching the customer is obtained based on a complex network community discovery model, and by analyzing the community structure and characteristics in the network, a comprehensive analysis and mining of the customer group is achieved.

[0057] Different from the problem of low recommendation accuracy in related technologies, the above method analyzes the customer's personal information indicators, financial status indicators, credit record indicators, behavioral characteristic indicators, personal preference indicators, loan history indicators, etc. when obtaining customer information, and conducts statistics and analysis on customer characteristics from various angles, thus realizing multi-dimensional analysis of customer information and personalized high-precision recommendations.

[0058] Different from the related art, which does not consider the difference of customer groups, the above method adopts a complex network community discovery model to fully consider the relationship and difference between customer groups.

[0059] The method in the embodiment of the present application can more accurately identify the needs and preferences of customers. By analyzing the relationships and interactions between customers, the method can divide customers into different communities or groups, each of which has similar characteristics and needs. This enables financial institutions to provide customers with personalized loan product recommendations based on the characteristics of different groups, thereby better meeting customer needs.

[0060] In one embodiment of the present application, the method of obtaining a loan product matched with the customer based on the customer information by using a complex network community discovery model includes: according to the customer information, taking the customer as a node in the network and taking the association relationship between customers as an edge in the network; dividing the nodes with similar characteristics in the network into different communities; and obtaining the similarity between customers in different communities through the complex network community discovery model as the loan product matched with the customer.

[0061] Complex network community discovery is an analysis method based on network structure, which can divide nodes with similar characteristics in the network into different communities. Based on the community discovery algorithm, a large number of customers can be quickly divided into different communities according to customer information, thereby achieving customer classification. When the network community structure is stable, the characteristics of each community are analyzed, including community density, global efficiency, node degree centrality and other indicators, to further explore the characteristics and rules within the community.

[0062] Specifically, customers can be regarded as nodes in the network, and the relationships between customers can be regarded as edges in the network. Then, the community discovery algorithm can be used to discover the similarities between customers, so as to recommend the most suitable products to customers. In the training stage of the model, a model is trained to discover the similarities between customers in different communities, so as to obtain loan products that can be matched with customers.

[0063] The method in the embodiment of the present application has high interpretability and reliability. By analyzing and modeling complex networks, the relationship and characteristics between different communities can be clearly displayed. This enables financial institutions to better understand the reasons and basis for the recommendation results, so as to be more confident in recommending suitable loan products to customers.

[0064] In one embodiment of the present application, obtaining a loan product matching the customer based on the customer information by using a complex network community discovery model also includes: counting customer information in different communities and the types of loan products historically selected by the customer to determine a proposed recommended loan product; and obtaining a loan product matching the customer based on the proposed recommended loan product.

[0065] By collecting statistics on customer information and the types of loan products they have historically selected under each submodule (the community contains multiple submodules), and calculating the relative importance of a certain loan product among customers in a submodule, we can obtain the preferred categories of theme loan products recommended by each submodule. In this way, when a customer needs a loan, we can determine the submodule to which he belongs based on his customer information, and then recommend theme loan products in the submodule.

[0066] The method in the embodiment of the present application makes recommendations based on the real relationships and behavior data between customers, and has high reliability and practicality.

[0067] In one embodiment of the present application, the obtaining of customer information in response to customer loan demands includes: in response to customer loan demands, using the Pearson correlation coefficient to calculate the correlation between each customer and construct a customer relationship network.

[0068] Through customer information, the Pearson correlation coefficient is used to calculate the correlation between customers and build a customer relationship network.

[0069] First, based on the above customer information, we need to express the relationship between customers using the Pearson correlation coefficient to get a number. Usually, a customer can be regarded as a node, and the line between each node is the weight between each two nodes. The graph formed is the network structure, that is, the customer relationship network.

[0070] Secondly, use two variables X and Y to represent the two customers, calculate the covariance and standard deviation between the two variables respectively, and the quotient between the two is the Pearson correlation coefficient between the two customers:

[0071]

[0072] By estimating the covariance and standard deviation of variables X and Y, we can get the Pearson correlation coefficient r between the two:

[0073]

[0074] To further optimize r in (2), we use a single variable (X i ,Y i ) is expressed as the mean of the standard scores, and the following expression is obtained:

[0075]

[0076] in Represents X i The average value, σ X Represents X i The standard deviation of Represents X i The standard score.

[0077] In one embodiment of the present application, the step of obtaining customer information in response to customer loan demand further includes: using a Louvain algorithm to perform community division on the customer relationship network to obtain a modular structure of the customer relationship network.

[0078] Specifically, the Louvain algorithm can be used to divide the customer relationship network into communities and obtain the modular structure of the customer relationship network:

[0079] On the basis of the customer relationship network, the modular structure of the customer relationship network is further explored, mainly through the Louvain algorithm in the community partition algorithm, to explore the structure of the customer relationship network.

[0080] The Louvain algorithm is mainly divided into two stages:

[0081] Phase 1: Number each node in the network from 1 to n. Each node can be regarded as a community, that is, a community number from 1 to n is obtained. For each node i, try to merge node i with its adjacent node j, and calculate the modularity increment ΔQ of the new module composed of these two nodes after merging. If ΔQ>0, the merge operation can be performed, that is, nodes i and j are combined into a new module; if ΔQ<0, the original independent structure of the two nodes is kept unchanged.

[0082] The formula for modularity increment ΔQ is as follows:

[0083]

[0084] By simplifying the above formula, we can get the following expression for the modularity increment:

[0085]

[0086] In the above formula, k i ,k jThey represent the degrees of nodes i and j respectively, and m is the total number of edges in the entire network.

[0087] Repeat the above process and continuously reorganize the adjacent nodes into modules until the module relationship of all nodes no longer changes. At this time, the first stage of module division is completed.

[0088] Phase 2: After the module reorganization in the first phase, a stable community structure is obtained. In this phase, each community as a whole is first treated as a new node, and then the nodes within each community are compressed. After the internal reorganization, the edges and nodes of the network have new connection relationships and weights. Finally, the first phase is repeated until a reorganized network has the largest local modularity.

[0089] The expression for calculating modularity is as follows:

[0090]

[0091] in k j is the degree of node j, is the probability that there is an edge between node j and each node in the network, is the number of edges between nodes i and j, A ij = 0 means that i and j are not connected, A ij =1 means that i and j are connected by an edge. C represents the community division of the node. The closer its value is to 1, the higher the division quality. Therefore, the expression of modularity Q can also be expressed as:

[0092]

[0093] Where ∑in is the sum of the weights of the edges inside module C, and ∑tot is the sum of the weights of the edges connecting the nodes inside module C.

[0094] Based on the above implementation principle, the specific steps of using the Louvain algorithm to divide the customer relationship network into communities are as follows:

[0095] S1: Number each node in the network from 1 to n. Each node can be regarded as a community, that is, the community number is 1 to n.

[0096] S2: For each node i, try to merge node i with its adjacent node j, and calculate the modularity increment ΔQ of the new module composed of the two nodes after merging. If ΔQ>0, the merge operation can be performed, that is, nodes i and j are combined into a new module; if ΔQ<0, the original independent structure of the two nodes is kept unchanged;

[0097] S3: Loop through S2 until the module relationships of all nodes do not change and then exit the loop;

[0098] S4: Compress the reorganized community structure. First, treat each community as a new node, and then compress the nodes within each community. Recalculate the connection weights between nodes within the community and mark them as the weights on the self-loop edges of the new nodes; the original connection weights between communities are marked as the connection weights between new nodes;

[0099] S5: Execute S1 and S2 repeatedly until the entire network structure is stable and the modularity does not change.

[0100] In one embodiment of the present application, the step of obtaining customer information in response to customer loan demand also includes: performing characteristic analysis on each community attribute and comparing the community attributes, wherein the community attributes include at least one of the following: global efficiency, local efficiency, clustering coefficient, average shortest path length, and node degree.

[0101] The characteristics of each community are analyzed, including community density, global efficiency, node degree centrality and other indicators, and the communities are compared.

[0102] After obtaining a stable community structure according to the community division algorithm, multiple sub-modules are included in the community structure, and the characteristics within each community are further analyzed, including indicators such as community density, global efficiency, and node degree centrality, and the communities are compared.

[0103] The following is a detailed description using global efficiency, local efficiency, clustering coefficient, and average shortest path length as examples, which is not intended to limit the protection scope of the embodiments of the present application.

[0104] Global efficiency refers to the efficiency of information transmission between any two nodes in the network. It is measured by calculating the average of the reciprocal of the shortest path length between all nodes in the network. It can reflect the overall information processing and transmission capabilities of the network and is an important indicator for measuring the overall performance of the network. It can also reflect the degree of integration of the network and help us evaluate and improve the network structure. ij is the distance between nodes i and j, and N is the number of network nodes. The expression of global efficiency is as follows:

[0105]

[0106] Local efficiency is an indicator used to measure the degree of network separation. It mainly focuses on the efficiency of information transmission between nodes in the network and their directly adjacent nodes, and can better reflect the processing and transmission capabilities of local information. Where N represents the total number of nodes in the network, and GE represents global efficiency. The expression of local efficiency is as follows:

[0107]

[0108] The clustering coefficient can quantify the closeness of the connection between adjacent nodes and is an important indicator to measure the degree of clustering of network nodes. Nodes with a higher degree of clustering have stronger capabilities in information transmission and collaborative cooperation. i is the clustering coefficient of node i, E i is the number of edges adjacent to the node i, k i Represents the total number of neighboring points of node i. The expression of clustering coefficient is as follows:

[0109]

[0110] The average shortest path length is an indicator used to measure the ability to transmit information between network nodes. When the average shortest path length in the network is low, it means that the efficiency of information transmission between nodes is high, and information can flow quickly between regions. This indicator can better reveal the level of functional integration between regions, and is of great significance for evaluating network performance, optimizing network structure to improve information transmission efficiency, and promoting collaborative cooperation between regions. ij is the distance between nodes i and j, and N is the number of network nodes. The expression of the average shortest path length is as follows:

[0111]

[0112] The node degree is composed of the sum of the out-degree and in-degree of the node, reflecting the number of connections between the node and other nodes. The core nodes and edge nodes in the network can be identified based on the node degree, which plays an important role in the study of network modules and community groups. The expression of node degree is as follows:

[0113]

[0114] The above global efficiency, local efficiency, clustering coefficient, average shortest path length and node degree by node need to be calculated and compared together when constructing a complex network community discovery model and when judging the similarity of customer information in the actual verification phase. If the results of global efficiency, local efficiency, clustering coefficient, average shortest path length and node degree by node are similar, it is considered to be related to the customer information, and the submodule where the customer is located is found in the network community.

[0115] In one embodiment of the present application, the method also includes: pre-acquiring personal information indicators, financial status indicators, credit record indicators, behavioral characteristic indicators, personal preference indicators, and loan history indicators; wherein the personal information indicators are used to reflect the customer's personal situation and stability, the financial status indicators are used to reflect the customer's financial situation and debt repayment ability, the credit record indicators are used to reflect the customer's credit record and credit risk, the behavioral characteristic indicators are used to reflect the customer's behavioral characteristics and lifestyle, the personal preference indicators are used to reflect the customer's personal preferences and needs, and the loan history indicators are used to reflect the customer's historical loan situation.

[0116] When training a complex network community discovery model, it is necessary to collect customer information and perform preprocessing, which mainly includes the following information:

[0117] (1) Personal information indicators: including but not limited to age, gender, marital status, education level, occupation, etc. These indicators can reflect the basic personal situation and stability of the customer.

[0118] (2) Financial status indicators: including but not limited to income, expenditure, deposits, liabilities, etc. These indicators can reflect the customer's financial status and debt repayment ability.

[0119] (3) Credit record indicators: including but not limited to credit rating, credit card usage, overdue status, etc. These indicators can reflect the customer's credit record and credit risk.

[0120] (4) Behavioral characteristic indicators: including but not limited to consumption habits, online shopping, social media usage, etc. These indicators can reflect customers’ behavioral characteristics and lifestyles.

[0121] (5) Personal preference indicators: including but not limited to investment preferences, consumption preferences, lifestyle, etc. These indicators can reflect the customer’s personal preferences and needs.

[0122] (6) Loan history indicators: including but not limited to the number of loan applications in the previous year, the total amount of loans in the previous year, overdue status, etc. These indicators can reflect the customer's historical loan status.

[0123] When loan product recommendations are needed, customer information is collected and pre-processed in real time, mainly including the following information: name, age, gender, marital status, education level, occupation, income, expenditure, deposits, liabilities, credit rating, consumption habits, investment preferences, historical loan information, etc.

[0124] By analyzing customers' personal information indicators, financial status indicators, credit record indicators, behavioral characteristic indicators, personal preference indicators, loan history indicators, etc., customer characteristics are counted and analyzed from various angles, realizing multi-dimensional analysis of customer information and personalized recommendations.

[0125] The present application embodiment also provides a loan product recommendation device 200, such as Figure 2 As shown, a schematic diagram of the structure of a loan product recommendation device in an embodiment of the present application is provided. The loan product recommendation device 200 at least includes: a response module 210 and a matching recommendation module 220, wherein:

[0126] In one embodiment of the present application, the response module 210 is specifically used to obtain customer information in response to customer loan demand.

[0127] Obtain user information based on customer loan needs. User information includes but is not limited to statistics and analysis results of customer characteristics from various angles to meet the needs of multi-dimensional customer information analysis and personalized recommendations. Specifically, it mainly includes the following information: name, age, gender, marital status, education level, occupation, income, expenditure, deposits, liabilities, credit rating, consumption habits, investment preferences, historical loan information, etc. In addition, it is necessary to pre-process the customer data in the customer loan demand to obtain customer information.

[0128] It should be noted that "customer loan demand" can be based on an Internet request or a local request.

[0129] In one embodiment of the present application, the matching recommendation module 220 is specifically used to: obtain a loan product matching the customer based on the customer information by using a complex network community discovery model, and the complex network community discovery model is used to discover similarities between customers.

[0130] Complex network community discovery technology is widely used in recommendation systems. Complex network community discovery is an analysis method based on network structure, which can divide nodes with similar characteristics in the network into different communities. The present invention regards customers as nodes in the network, and the associations between customers as edges in the network, and then uses the community discovery algorithm to discover the similarities between customers, so as to recommend the most suitable products to customers.

[0131] Based on the acquired customer information, the complex network community discovery model is used to match the loan products. The complex network community discovery model can find matching loan products by discovering the similarities between customers. When a customer needs a loan, the community to which the customer belongs is determined based on the customer's customer information, and then the theme loan products in the community are recommended.

[0132] In one embodiment of the present application, the matching recommendation module 220 is also used to

[0133] According to the customer information, the customers are regarded as nodes in a network and the associations between customers are regarded as edges in the network;

[0134] Divide nodes with similar characteristics in the network into different communities;

[0135] The complex network community discovery model is used to obtain similarities between customers in different communities, which is used to match loan products with customers.

[0136] In one embodiment of the present application, the matching recommendation module 220 is also used to

[0137] Collect customer information in different communities and the types of loan products that customers have historically selected, and determine the loan products to be recommended;

[0138] Based on the proposed recommended loan product, a loan product matching the customer is obtained.

[0139] In one embodiment of the present application, the response module 210 is further configured to:

[0140] In response to customer loan demands, the Pearson correlation coefficient is used to calculate the correlation between customers and construct a customer relationship network.

[0141] In one embodiment of the present application, the response module 210 is further configured to:

[0142] The Louvain algorithm is used to divide the customer relationship network into communities, thereby obtaining a modular structure of the customer relationship network.

[0143] In one embodiment of the present application, the response module 210 is further configured to:

[0144] A characteristic analysis is performed on each community attribute, and the community attributes are compared, wherein the community attributes include at least one of the following: global efficiency, local efficiency, clustering coefficient, average shortest path length, and node degree.

[0145] It can be understood that the above-mentioned loan product recommendation device can implement each step of the loan product recommendation method provided in the above-mentioned embodiment, and the relevant explanations about the loan product recommendation method are applicable to the loan product recommendation device and will not be repeated here.

[0146] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 3At the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. The memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0147] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0148] The memory is used to store the program. Specifically, the program may include a program code, and the program code includes a computer operation instruction. The memory may include a memory and a non-volatile memory, and provides instructions and data to the processor.

[0149] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a loan product recommendation device at the logical level. The processor executes the program stored in the memory and is specifically used to perform the following operations:

[0150] Respond to customer loan requests and obtain customer information;

[0151] According to the customer information, a loan product matching the customer is obtained based on a complex network community discovery model, and the complex network community discovery model is used to discover similarities between customers.

[0152] The above application Figure 1The method performed by the loan product recommendation device disclosed in the illustrated embodiment can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by a hardware integrated logic circuit in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiments of the present application can be directly embodied as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0153] The electronic device may also perform Figure 1 The method executed by the loan product recommendation device in Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0154] The present application also provides a computer-readable storage medium, which stores one or more programs, wherein the one or more programs include instructions, which, when executed by an electronic device including multiple application programs, enable the electronic device to execute Figure 1 The method performed by the loan product recommendation device in the illustrated embodiment is specifically used to perform:

[0155] Respond to customer loan requests and obtain customer information;

[0156] According to the customer information, a loan product matching the customer is obtained based on a complex network community discovery model, and the complex network community discovery model is used to discover similarities between customers.

[0157] 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.

[0158] 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.

[0159] 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.

[0160] 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.

[0161] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0162] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0163] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0164] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0165] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt 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.) that contain computer-usable program code.

[0166] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A loan product recommendation method, wherein: The recommended methods include: Respond to customer loan requests and obtain customer information; According to the customer information, a loan product matching the customer is obtained based on a complex network community discovery model, and the complex network community discovery model is used to discover similarities between customers.

2. The method of claim 1, wherein: The method of obtaining a loan product matching the customer based on the customer information by using a complex network community discovery model includes: According to the customer information, the customers are regarded as nodes in a network and the associations between customers are regarded as edges in the network; Divide nodes with similar characteristics in the network into different communities; The complex network community discovery model is used to obtain similarities between customers in different communities, which is used to match loan products with customers.

3. The method of claim 2, wherein: The method of obtaining a loan product matching the customer based on the customer information by using a complex network community discovery model also includes: Collect customer information in different communities and the types of loan products that customers have historically selected, and determine the loan products to be recommended; Based on the proposed recommended loan product, a loan product matching the customer is obtained.

4. The method of claim 1, wherein: The step of obtaining customer information in response to a customer's loan demand includes: In response to customer loan demands, the Pearson correlation coefficient is used to calculate the correlation between customers and construct a customer relationship network.

5. The method of claim 4, wherein: The step of obtaining customer information in response to the customer's loan demand may further include: The Louvain algorithm is used to divide the customer relationship network into communities, thereby obtaining a modular structure of the customer relationship network.

6. The method of claim 4, wherein: The step of obtaining customer information in response to the customer's loan demand may further include: A characteristic analysis is performed on each community attribute, and the community attributes are compared, wherein the community attributes include at least one of the following: global efficiency, local efficiency, clustering coefficient, average shortest path length, and node degree.

7. The method according to any one of claims 1 to 6, further comprising: Pre-acquire personal information indicators, financial status indicators, credit record indicators, behavioral characteristic indicators, personal preference indicators, and loan history indicators; in, The personal information indicators are used to reflect the customer's personal situation and stability. The financial status indicators are used to reflect the financial status and debt repayment ability of the customer. The credit record indicator is used to reflect the customer's credit record and credit risk. The behavioral characteristic indicators are used to reflect the behavioral characteristics and lifestyle of customers. The personal preference index is used to reflect the personal preferences and needs of customers. The loan history indicator is used to reflect the customer's historical loan situation.

8. A loan product recommendation device, wherein: The device comprises: A response module, used to respond to customer loan demands and obtain customer information; The matching recommendation module is used to obtain loan products that match the customer based on the customer information through a complex network community discovery method, and the complex network community discovery model is used to discover similarities between customers.

9. An electronic device, comprising: processor; as well as A memory arranged to store computer executable instructions, which when executed cause the processor to perform the method of any one of claims 1 to 7.

10. A computer-readable storage medium storing one or more programs, which, when executed by an electronic device including a plurality of application programs, causes the electronic device to execute any one of the methods of claims 1 to 7.

Images