Method and device for identifying candidate object

Abstract

The embodiment of the invention provides a candidate object identification method and device, and the method comprises the steps: obtaining a first object set which comprises a plurality of target objects, and the credit of the target objects is higher than a preset credit threshold value; a to-be-recognized object is obtained, the similarity between the to-be-recognized object and each target object in the first object set is determined, and the to-be-recognized object is other objects except the object in the first object set; and determining whether the to-be-identified object is a candidate object or not according to the similarity. Through the method and the device, the problem that potential customers are difficult to accurately identify in related technologies is solved, and the effect of improving the success rate of business handling is further achieved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computers, and in particular, to a method and device for identifying candidate objects. Background Art

[0002] Business development is a key part of company development. Through business development, companies can improve their competitiveness in a highly competitive market environment. The core of business development is to find potential customers.

[0003] Taking banks as an example, corporate customer development is the core of banking business development and an important carrier for banks to obtain profits and maintain relationships with banks. The core of corporate customer development is to tap into the value of a huge number of potential customers and build customer relationships, thereby enhancing the bank's brand image and increasing the bank's business income. However, how to tap into potential customers with similar characteristics to the bank's high-quality customers from a large number of potential customers and provide them with valuable potential customer recommendation strategies is a key issue for banks to expand corporate customers.

[0004] There is currently no effective solution to the above problems. Summary of the invention

[0005] The embodiments of the present invention provide a method and device for identifying candidate objects, so as to at least solve the problem in the related art that it is difficult to accurately identify potential customers.

[0006] According to one embodiment of the present invention, there is provided a method for identifying candidate objects, comprising: obtaining a first object set, wherein the first object set includes a plurality of target objects, and the credit of the target objects is higher than a preset credit threshold; obtaining an object to be identified, and determining a similarity between the object to be identified and each target object in the first object set, wherein the object to be identified is an object other than the objects in the first object set; and determining whether the object to be identified is a candidate object based on the similarity.

[0007] In an exemplary embodiment, determining whether the object to be identified is a candidate object based on the similarity includes: determining a target credit value of the object to be identified when there is an object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold; determining whether the object to be identified is the candidate object based on the target credit value; and determining that the object to be identified is not the candidate object when there is no object in the first object set whose similarity with the object to be identified is greater than the preset similarity threshold.

[0008] In an exemplary embodiment, determining the target credit value of the object to be identified includes: determining the target object that has an association relationship with the object to be identified in the first object set to obtain a first object subset; determining multiple credit parameters of the object to be identified based on the credit values ​​of each target object in the first object subset to obtain a credit parameter set; and determining the target credit value of the object to be identified based on each credit parameter in the credit parameter set.

[0009] In an exemplary embodiment, the target credit value of the object to be identified is determined based on each credit parameter in the credit parameter set, including: determining the weighted sum of each credit parameter in the credit parameter set as a first target sum; determining the sum of each credit parameter in the credit value set as a second target sum; and determining the ratio of the first target sum to the second target sum as the target credit value of the object to be identified.

[0010] In an exemplary embodiment, multiple credit parameters of the object to be identified are determined based on the credit values ​​of each of the target objects in the first object subset, including: obtaining associated objects of the first target object in the first object subset to obtain a set of associated objects of the first target object, wherein the first target object is any target object in the first object subset, the associated object is an object that has the association relationship with the first target object, and the associated object set includes the object to be identified; determining the number of associated objects in the associated object set to obtain a first quantity value; determining a first credit parameter of the object to be identified through the first quantity value and the credit value of the first target object, wherein the credit parameter set includes the first credit parameter.

[0011] In an exemplary embodiment, determining the first credit parameter of the object to be identified by using the first quantity value and the credit value of the first target object includes: obtaining the ranking of the object to be identified in the associated object set to obtain the ranking value of the object to be identified, wherein each object in the associated object set is sorted according to the degree of association with the first target object; determining the first credit value of the object to be identified based on the credit value of the first target object, the first quantity and the ranking value.

[0012] In an exemplary embodiment, determining the first credit parameter of the object to be identified based on the credit value of the first target object, the first quantity and the ranking value includes: obtaining a target parameter value based on the first quantity and the ranking value; and determining the product of the target parameter value and the credit value of the first target object as the first credit parameter.

[0013] In an exemplary embodiment, after determining whether the object to be identified is a candidate object based on the similarity, the method further includes: if the object to be identified is the candidate object, assigning a first identifier to the candidate object, wherein the target object in the first object set is assigned a second identifier.

[0014] According to another embodiment of the present invention, there is provided a device for identifying candidate objects, comprising: a first acquisition module, used to acquire a first object set, wherein the first object set includes multiple target objects, and the credit of the target objects is higher than a preset credit threshold; a second acquisition module, used to acquire an object to be identified, and determine the similarity between the object to be identified and each target object in the first object set, wherein the object to be identified is an object other than the objects in the first object set; and an identification module, used to determine whether the object to be identified is a candidate object based on the similarity.

[0015] According to yet another embodiment of the present invention, a computer-readable storage medium is provided, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program implements the steps of any of the above methods when executed by a processor.

[0016] According to yet another embodiment of the present invention, there is provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0017] According to yet another embodiment of the present invention, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0018] Through the present invention, a target object whose internal credit is higher than a preset credit threshold is obtained to obtain a first object set, and based on the similarity between the object to be identified and each target object in the first object set, it is determined whether the object to be identified is a candidate object.

[0019] By developing new customers and identifying and expanding valuable customers around existing high-quality customers, a list of high-quality creditable customers is generated, which can reduce the work of grassroots account managers in contacting and expanding customers to the greatest extent and increase the probability of contacting customers and granting credit. Therefore, the problem of difficulty in accurately identifying potential customers in related technologies can be solved, achieving the effect of improving the success rate of business handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a hardware structure block diagram of a mobile terminal according to a method for identifying candidate objects according to an embodiment of the present invention;

[0021] Figure 2 is a flowchart of a method for identifying candidate objects according to an embodiment of the present invention;

[0022] Figure 3 is a system logic diagram for identifying potential customers in a banking system according to an embodiment of the present invention;

[0023] Figure 4 is a flow chart of customer graph construction according to an embodiment of the present invention;

[0024] Figure 5 is a flowchart of potential customer mining according to an embodiment of the present invention;

[0025] Figure 6 4 is a structural block diagram of an apparatus for identifying candidate objects according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings and in combination with the embodiments.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0028] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 FIG. 1 is a hardware structure block diagram of a mobile terminal of a method for identifying candidate objects according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0029] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the method for identifying candidate objects in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0030] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0031] In this embodiment, a method for identifying candidate objects running on the above mobile terminal or network architecture is provided. Figure 2 is a flowchart of a method for identifying candidate objects according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0032] Step S202, obtaining a first object set, wherein the first object set includes a plurality of target objects, and the credit of the target objects is higher than a preset credit threshold;

[0033] The above-mentioned target objects can be customers screened out from existing customers based on actual needs, such as high-quality customers (i.e., customers whose credit is higher than a preset credit threshold) among customers who have currently participated in or purchased a certain business, such as customers who have applied for loans or credit cards within a bank, customers who have purchased certain goods in a store, customers who have purchased a certain insurance within an insurance company, etc. Through the target objects, the system can mine potential customers that meet actual needs based on the similarity between other objects and the target objects; the above-mentioned first object set can be a set of target objects, which can include some or all target objects. It should be noted that the above-mentioned method for confirming high-quality customers is only a preferred embodiment, and other methods can be selected to determine high-quality customers (i.e., target objects) according to actual conditions.

[0034] For example, when a bank needs to tap into potential customers who are applying for loans, it uses the customer's credit rating or creditworthiness as an entry point. At this time, it is necessary to obtain the credit (i.e., credit value) of all customers who have already applied for loans within the bank, and select customers whose credit is higher than a preset threshold as target objects, thereby obtaining a set of target objects (i.e., a first object set).

[0035] Step S204, obtaining an object to be identified, and determining the similarity between the object to be identified and each target object in the first object set, wherein the object to be identified is an object other than the objects in the first object set;

[0036] The above-mentioned object to be identified may be an object that needs to be confirmed as a potential customer. The object to be identified may be any internal or external customer who has not participated in or purchased a certain business, such as a customer who has not applied for a credit card or loan within the bank, or a customer outside the bank, a customer who has consumed in the store but has not purchased a certain product, or a customer who has not consumed in the store, etc.; the above-mentioned similarity may be the degree of similarity or correlation between the target object and the object to be identified. Through the similarity between the two, the difference between the two objects in some aspects or fields can be determined, wherein the greater the similarity, the greater the degree of similarity between the object to be identified and the target object.

[0037] The object to be identified is obtained, and the object to be identified is matched and compared with each target object in the first object set to determine the similarity between the object to be identified and each target object, so as to determine whether the object to be identified is a potential customer (ie, a candidate object) based on the similarity between the two.

[0038] As an optional implementation, the similarity between the object to be identified and the target object can be determined based on the customer's attribute information. For example, semantic labels are assigned to each object to be identified and the target object, such as customer type, industry type, customer size code, etc., and the specific attribute information corresponding to each semantic label is confirmed according to the actual situation of the object, so as to determine the data vector of the object according to the attribute information of the object, and the data vector of the object to be identified and the data vector of the target object are used to calculate the similarity between the object to be identified and the target object. Since the attribute overlap between customer nodes is proportional to the actual situation between customers, for example, the higher the degree of overlap of customer attributes, the more similar the business status and enterprise development trend are. Under the same environment, the two may face similar capital turnover situations, etc. Therefore, by determining the similarity between customers through the customer's attribute information, the overall characteristics and behavioral habits of the customer group can be better understood, thereby improving the accuracy of expanding business and services. At the same time, it can also help enterprises better formulate product positioning and market strategies to meet the needs of different customer groups and increase sales and customer loyalty.

[0039] Optionally, the data vector of the customer (ie, the target object or the object to be identified) can be obtained according to the following formula:

[0040]

[0041] In the formula, sen vec is the data vector of the above object, which can be the target object or the object to be identified. i is the vector of the above attribute information, which can be obtained by converting each attribute information through the word2vec word vector conversion tool. n is the number of attribute information that needs to be compared. For example, the target object and the object to be identified both include 5 attribute information, but only 3 of them are used to match and compare the similarity between the two, then n is 3.

[0042] Optionally, the similarity between the target object and the object to be identified can be obtained by the following formula:

[0043]

[0044] Where s(x,y) 2 is the similarity between the target object and the object to be identified, v(x,y) 2 is the difference between the target object and the object to be identified, x is the data vector of the target object, y is the data vector of the object to be identified, x i is the vector of attribute information of the target object, y i is the vector of attribute information of the above-mentioned object to be identified, and n is the number of attribute information that needs to be compared.

[0045] Step S206: determining whether the object to be identified is a candidate object according to the similarity.

[0046] Optionally, determining whether the object to be identified is a candidate object based on the similarity includes: determining a target credit value of the object to be identified when there is an object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold; determining whether the object to be identified is the candidate object based on the target credit value; and determining that the object to be identified is not the candidate object when there is no object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold.

[0047] The target credit value can be the credit value of the object to be identified; when it is determined that there is no object in the first object set whose similarity with the object to be identified is greater than the preset similarity threshold, it means that the difference between the object to be identified and the target object is large, that is, the correlation is low, and it is determined that the object to be identified is not a candidate object or a potential customer; when it is determined that there is an object in the first object set whose similarity with the object to be identified is greater than the preset similarity threshold, it means that the difference between the object to be identified and the target object is small, that is, the correlation is high, and it is determined that the object to be identified can be used as an alternative candidate object or a potential customer. At this time, it is necessary to determine the target credit value of the object to be identified, so as to finally determine whether the object to be identified is a candidate object or a potential customer based on the target credit value of the object to be identified. By jointly identifying candidate objects or potential customers through the similarity with existing high-quality customers and the credit value of the subset of objects to be identified, relatively high-quality customers can be screened out from potential customers, thereby helping enterprises to better understand and manage potential customer groups, so as to improve the overall quality of customers and customer conversion rate under the premise of ensuring risk control.

[0048] Optionally, the execution subject of the above steps can be a background processor, or other devices with similar processing capabilities, or a machine that integrates at least an image acquisition device and a data processing device, wherein the image acquisition device may include a graphics acquisition module such as a camera, and the data processing device may include a computer, a mobile phone or other terminal, but is not limited to this.

[0049] Through the above steps, the target objects whose internal credit is higher than the preset credit threshold are obtained to obtain the first object set, and based on the similarity between the object to be identified and each target object in the first object set, it is determined whether the object to be identified is a candidate object, so as to carry out customer mining and valuable customer identification and expansion around the existing high-quality customers, thereby generating a list of high-quality creditable customers, which can reduce the work of grassroots account managers in contacting and expanding customers to the greatest extent and increase the probability of contacting customers to credit. It solves the problem of difficulty in accurately identifying potential customers in related technologies and improves the success rate of business handling.

[0050] As an optional implementation, determining the target credit value of the object to be identified includes: determining the target object that has an association relationship with the object to be identified in the first object set to obtain a first object subset; determining multiple credit parameters of the object to be identified based on the credit values ​​of each target object in the first object subset to obtain a credit parameter set; and determining the target credit value of the object to be identified based on each credit parameter in the credit parameter set.

[0051] The above-mentioned association relationship can be a transaction relationship between customers. For example, in a banking system, the association relationship can be a transaction relationship between customers. In a store, the association relationship can be a friendship relationship between customers. In an e-commerce system, the association relationship can be a relationship of browsing the same goods, etc.; all target objects that have an association relationship with the object to be identified are determined in the first object set to obtain a first object subset, and based on the credit values ​​of each target object in the first object subset, multiple credit parameters of the object to be identified are obtained, so as to finally determine the target credit value of the object to be identified based on the multiple credit parameters, thereby ensuring the accuracy of the credit value of the object to be identified.

[0052] For example, if the object A to be identified has an association relationship with the target objects B, C, and D in the first object set, then the first object subset is determined to be (B, C, D), and the credit parameter t1 of the object A to be identified is determined according to the credit value of the target object B, the credit parameter t2 of the object A to be identified is determined according to the credit value of the target object C, and the credit parameter t3 of the object A to be identified is determined according to the credit value of the target object D, thereby determining the target credit value T of the object to be identified according to the credit parameters t1, t2, and t3.

[0053] As an optional implementation, the target credit value of the object to be identified is determined based on each credit parameter in the credit parameter set, including: determining the weighted sum of each credit parameter in the credit parameter set as a first target sum; determining the sum of each credit parameter in the credit parameter set as a second target sum; and determining the ratio of the first target sum to the second target sum as the target credit value of the object to be identified.

[0054] Optionally, the target credit value of the object to be identified may be determined by the following formula:

[0055]

[0056] Where T is the target credit value, t1, t2, t m are the above credit parameters, w1, w2, w m is the preset weight.

[0057] As an optional implementation, multiple credit parameters of the object to be identified are determined based on the credit values ​​of each of the target objects in the first object subset, including: obtaining associated objects of the first target object in the first object subset to obtain a set of associated objects of the first target object, wherein the first target object is any target object in the first object subset, the associated object is an object that has the association relationship with the first target object, and the associated object set includes the object to be identified; determining the number of associated objects in the associated object set to obtain a first quantity value; determining the first credit parameter of the object to be identified through the first quantity value and the credit value of the first target object, wherein the credit parameter set includes the first credit parameter.

[0058] The first target object may be any target object in the first object subset; the first associated object set may be a set of objects that are associated with the first target object, and the associated object set includes the object to be identified, wherein the objects in the associated object set may be all objects that are associated with the first target object, or may be some objects that are associated with the first target object, and the some objects may be objects that meet certain preset conditions among the objects that are associated with the first target object, for example, objects that have a transaction relationship with the first target object and have a transaction number greater than or equal to 10 times and an amount greater than RMB 1 million in the past 6 months; objects that are associated with the first target object (the objects may be internal or external) are obtained to obtain an associated object set, and the number of objects in the associated object set is determined to obtain a first quantity value, thereby determining a first credit parameter of the object to be identified based on the first quantity value and the credit value of the first target object.

[0059] As an optional implementation, determining the first credit parameter of the object to be identified through the first quantity value and the credit value of the first target object includes: obtaining the ranking of the object to be identified in the associated object set to obtain the ranking value of the object to be identified, wherein each object in the associated object set is sorted according to the degree of association with the first target object; determining the first credit value of the object to be identified based on the credit value of the first target object, the first quantity and the ranking value.

[0060] The above ranking value can be the ranking of the object to be identified in the set of associated objects, wherein the specific ranking method can be carried out with respect to the degree of association between the object to be identified and the first target object; the ranking value of the object to be identified in the set of associated objects is obtained, and then the first credit value of the object to be identified is determined according to the credit value, the first quantity and the ranking value of the first target object.

[0061] Optionally, the above-mentioned degree of association may be represented by a relationship importance, wherein the greater the relationship importance, the greater the degree of association, and the higher the ranking. For example, the relationship importance may be calculated by the following formula:

[0062] P=a×l+(1-l)×b

[0063] Where P is the importance of the above relationship, a is the number of transactions with the first target object, b is the average transaction amount with the first target object, and l is the weight factor, which takes a value between [0,1].

[0064] It should be noted that the above relationship importance is only a preferred embodiment, and the degree of association with the first target object may also be determined by other indicators.

[0065] As an optional implementation, determining the first credit parameter of the object to be identified based on the credit value of the first target object, the first quantity and the ranking value includes: obtaining a target parameter value based on the first quantity and the ranking value; and determining the product of the target parameter value and the credit value of the first target object as the first credit parameter.

[0066] Optionally, the first credit parameter of the object to be identified may be determined by the following formula:

[0067]

[0068] In the formula, t m is the first credit parameter, sco is the credit value of the first target object, R is the first quantity value, r is the ranking value, c is the attenuation coefficient, is the target parameter value mentioned above.

[0069] It should be noted that the above-mentioned method for calculating the target parameter value is only a preferred method, and the calculation can be specifically set according to the actual situation.

[0070] As an optional implementation, determining whether the object to be identified is the candidate object based on the credit value includes: when the credit value is greater than or equal to a preset threshold, determining that the object to be identified is the candidate object; when the credit value is less than the preset threshold, determining that the object to be identified is not the candidate object.

[0071] As an optional implementation, after determining whether the object to be identified is a candidate object based on the similarity, the method further includes: if the object to be identified is the candidate object, assigning a first identifier to the candidate object, wherein the target object in the first object set is assigned a second identifier.

[0072] The above-mentioned first identifier and second identifier can be labels used to represent different customers. After determining the candidate objects, different identifiers are assigned to the target objects and the candidate objects respectively, so as to effectively distinguish between existing high-quality customers and potential customers, and can more accurately classify and locate customers, which is conducive to formulating personalized business expansion strategies and promotion plans for different objects, and providing strong support for business expansion work.

[0073] As an optional implementation, Figure 3 is a system logic diagram for identifying potential customers in a banking system according to an embodiment of the present invention, such as Figure 3 As shown, specifically including the following:

[0074] S1, corporate customer data collection;

[0075] By focusing on internal data mining and introducing external data, we have established a comprehensive, visual, and infinitely expandable customer relationship network map, realizing the transformation of public-private linkage and circle-chain customer development to a new model, empowering customers, managers, and customer development personnel. The specific customer data is as follows:

[0076] (1) Basic information of the customer (i.e., the object), including customer keywords, customer number, customer basic account opening mark, non-performing asset write-off customer mark, customer size code, customer opening institution, customer type, industry type, organization code, enterprise three-in-one certificate number, customer credit rating, etc.

[0077] (2) Intra-bank customer transaction details: sort out the internal connections and transaction exchanges between our customers and customers of other banks, and obtain specific transaction details. For example, if there are multiple transactions between customer A and customer B within a certain period of time, obtain the account nature and transaction currency of customer A and customer B, as well as the total transaction funds, number of funds, and average transaction amount (per transaction) under the combination of account nature and number of transactions, such as customer A's total outflow amount (converted into RMB), customer A's total inflow amount (converted into RMB), customer A's total number of outflow funds, and customer A's total number of outflow funds.

[0078] (3) Enterprise public information. The National Financing Credit Service Platform can share national enterprise financing credit information and credit services, provide credit financing services to small, medium and micro enterprises, and feedback financing results information to the National Financing Credit Service Platform. The Financing Credit Service Platform contains 26 types of enterprise information, including market entity registration information, judicial information, tax information, housing provident fund information, social insurance information, ecological environment information, import and export information, intellectual property information, software copyright information, real estate information, administrative management information, water, electricity and gas fee payment information, scientific and technological research and development information, special and innovative enterprise certification information, financial funding project information, demonstration information, guarantee information, new agricultural business entity information, enterprise serious dishonesty punishment list information, natural person serious dishonesty punishment list information, public resource transaction information, medical insurance designated private medical institutions information, agricultural-related list information, existing enterprise registration information, enterprise qualification certificate information, and enterprise honors and commendations information.

[0079] (4) Unstructured news opportunities: Banks can use the financial big model to generate news summaries and business opportunity objects in the news based on the content of external news, so as to timely and comprehensively understand the investment intentions, loan intentions, supply chain-related and other business opportunity information of enterprises. In addition, for corporate long-tail customers, there will be customer micro-product consultation records and customer online behaviors, including the various financial products they have viewed, the time they stay on the page, the frequency of clicks, etc. These data can help us depict the customer's interest map and demand tendency.

[0080] S2, construction of corporate-level bank customer map;

[0081] S21, data standardization processing;

[0082] Through data cleaning, data that does not meet the calculation requirements is filtered, and data is filtered, filled, deduplicated, etc. In the case of incomplete data, if it cannot be matched with the existing data according to its characteristics, it can be completed by the original system.

[0083] At the same time, data standardization is used to remove spaces from source data, convert data types, unify calculated values, etc. Among them, it is necessary to unify the granularity of the calculation fields involved. For example, the transaction amount can be uniformly converted into "RMB conversion" according to the exchange rate table.

[0084] S22, entity recognition and relation extraction;

[0085] The structured data of corporate customers is divided into multiple system data. Here, both internal and external data use the enterprise unified credit code as the primary key ID. After matching the enterprise information table according to the enterprise name, the unstructured data is appended with the enterprise unified credit code as a description field, which serves as a bridge for the integration of structured and unstructured data.

[0086] S23, assigning semantic labels;

[0087] Assign semantic labels to each customer, such as customer type, industry type, customer size code, etc., and obtain attribute information corresponding to each semantic label based on the actual situation of the customer.

[0088] S24, building an enterprise-level customer knowledge graph;

[0089] In the bank's enterprise-level customer knowledge graph, entities are mainly customer nodes, which are uniquely specified by the enterprise's unified credit code after hash coding. There are relationships such as investment, guarantee, transaction, and supply chain between customers. Multiple edges are defined in the graph to describe the relationship between customers. The present invention uses the Nebula graph database to store and identify the bank's enterprise-level knowledge graph, which is convenient for unified entry query, management, and update. Avoid data redundancy, data islands, and information inconsistency.

[0090] The specific construction process is as follows Figure 4 shown.

[0091] S3, potential customer mining based on enterprise-level graphs;

[0092] The word2vec word vector conversion tool is used to convert the word vector of the customer node attributes. The conversion method is as follows:

[0093]

[0094] In the formula, sen vec is the data vector of the above object (i.e., customer or customer node), which can be the target object or the object to be identified. i is the vector of the above attribute information, n is the number of attribute information that needs to be compared, for example, the target object and the object to be identified both include 5 attribute information, but only 3 of the attribute information are used to match and compare the similarity between the two, then n is 3.

[0095] At the same time, according to the data vector, the distance value of the data vector of the client node is calculated to generate the client node difference (that is, the difference between the target object and the object to be identified), and the similarity (that is, the similarity between the target object and the object to be identified) is obtained according to the difference. The acquisition method is as follows:

[0096]

[0097] Where s(x,y) 2 is the similarity between the target object and the object to be identified, v(x,y) 2 is the difference between the target object and the object to be identified, x is the data vector of the target object, y is the data vector of the object to be identified, x i is the vector of attribute information of the target object, y i is the vector of attribute information of the above-mentioned object to be identified, and n is the number of attribute information that needs to be compared.

[0098] Finally, according to the similarity (which can be converted into difference), a list of recommended customers is generated based on the preset threshold coefficient (i.e. the preset similarity threshold). The specific process is as follows: Figure 5 shown.

[0099] S4 Identification of potential valuable customers;

[0100] Randomly select an object to be identified (or customer node) in the customer list, obtain target locations that have an associated relationship (which may be a transaction relationship) with the object to be identified in the high-quality customer nodes (i.e., the first object set), and obtain a first object subset;

[0101] The following operations are performed on any object in the first object subset (ie, the first target object):

[0102] Obtaining a set of associated objects that have an associated relationship with the first target object, wherein the associated object set includes the object to be identified, and the objects in the associated object set meet preset rules, for example: within the past 6 months, the number of transactions with the first target is greater than or equal to 10 times, and the amount is greater than 1 million yuan, etc.;

[0103] All objects in the associated object set are sorted according to the associated importance customers. The relationship importance is as follows:

[0104] P=a×l+(1-l)×b

[0105] Where P is the importance of the above relationship, a is the number of transactions with the first target object, b is the average transaction amount with the first target object, and l is the weight factor, which takes a value between [0,1].

[0106] At the same time, the credit parameters of the object to be identified are determined according to the sorting results, as follows:

[0107]

[0108] Where, t m is the first credit parameter, sco is the credit value of the first target object, R is the first quantity value, r is the ranking value, c is the attenuation coefficient, is the target parameter value mentioned above.

[0109] Finally, multiple credit parameters of the object to be identified are obtained, and the target credit value of the object to be identified is obtained:

[0110]

[0111] Where T is the target credit value, t1, t2, t m are the above credit parameters, w1, w2, w m is the preset weight.

[0112] Determine whether the object to be identified is a potential customer (i.e., candidate) based on the target credit value. By building a graph between the bank's high-quality customers and their close trading partners, we can identify high-quality uncredited customers within the bank's core customer relationship circle, and achieve accurate, efficient, and high-quality customer acquisition.

[0113] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0114] In the present embodiment, a device for identifying candidate objects is also provided, and the device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made are not repeated. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware is also possible and conceived.

[0115] Figure 6 is a structural block diagram of an apparatus for identifying candidate objects according to an embodiment of the present invention, such as Figure 6As shown, the device includes: a first acquisition module 602, used to acquire a first object set, wherein the first object set includes multiple target objects, and the credit of the target object is higher than a preset credit threshold; a second acquisition module 604, used to acquire an object to be identified, and determine the similarity between the object to be identified and each target object in the first object set, wherein the object to be identified is other objects except the objects in the first object set; an identification module 606, used to determine whether the object to be identified is a candidate object based on the similarity.

[0116] In an exemplary embodiment, the device is also used to determine a target credit value of the object to be identified when there is an object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold; determine whether the object to be identified is the candidate object based on the target credit value; and determine that the object to be identified is not the candidate object when there is no object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold.

[0117] In an exemplary embodiment, the device is also used to determine the target object that has an association relationship with the object to be identified in the first object set to obtain a first object subset; determine multiple credit parameters of the object to be identified based on the credit values ​​of each target object in the first object subset to obtain a credit parameter set; determine the target credit value of the object to be identified based on each credit parameter in the credit parameter set.

[0118] In an exemplary embodiment, the device is also used to determine the weighted sum of each credit parameter in the credit parameter set as a first target sum; determine the sum of each credit parameter in the credit parameter set as a second target sum; and determine the ratio of the first target sum to the second target sum as the target credit value of the object to be identified.

[0119] In an exemplary embodiment, the device is also used to obtain associated objects of the first target object in the first object subset, and obtain an associated object set of the first target object, wherein the first target object is any target object in the first object subset, the associated object is an object that has the association relationship with the first target object, and the associated object set includes the object to be identified; determine the number of the associated objects in the associated object set to obtain a first quantity value; determine a first credit parameter of the object to be identified through the first quantity value and the credit value of the first target object, wherein the credit parameter set includes the first credit parameter.

[0120] In an exemplary embodiment, the device is also used to obtain the ranking of the object to be identified in the associated object set, and obtain the ranking value of the object to be identified, wherein each object in the associated object set is sorted according to the degree of association with the first target object; and determine the first credit value of the object to be identified based on the credit value of the first target object, the first quantity and the ranking value.

[0121] In an exemplary embodiment, the device is further configured to obtain a target parameter value according to the first quantity and the ranking value; and determine the product of the target parameter value and the credit value of the first target object as the first credit parameter.

[0122] In an exemplary embodiment, the apparatus is further configured to, when the object to be identified is the candidate object, assign a first identifier to the candidate object, wherein the target object in the first object set is assigned a second identifier.

[0123] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0124] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0125] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0126] An embodiment of the present invention further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0127] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0128] An embodiment of the present invention further provides a computer program product, including a computer program, which implements the steps of the method described in each embodiment of the present application when executed by a processor.

[0129] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0130] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0131] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying a candidate object, characterized in that include: Acquire a first object set, wherein the first object set includes a plurality of target objects, and the credit of the target objects is higher than a preset credit threshold; Acquire an object to be identified, and determine the similarity between the object to be identified and each target object in the first object set, wherein the object to be identified is an object other than the objects in the first object set; Determine whether the object to be identified is a candidate object according to the similarity.

2. The method according to claim 1, characterized in that Determining whether the object to be identified is a candidate object according to the similarity includes: If there is an object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold, determining a target credit value of the object to be identified; and determining whether the object to be identified is the candidate object according to the target credit value; When there is no object in the first object set whose similarity with the object to be identified is greater than a preset similarity threshold, it is determined that the object to be identified is not the candidate object.

3. The method according to claim 2, characterized in that Determining a target credit value of the object to be identified includes: Determine the target object that has an association relationship with the object to be identified in the first object set to obtain a first object subset; Determining multiple credit parameters of the to-be-identified object according to the credit values ​​of each of the target objects in the first object subset to obtain a credit parameter set; The target credit value of the object to be identified is determined according to each credit parameter in the credit parameter set.

4. The method according to claim 3, characterized in that Determining the target credit value of the object to be identified according to each credit parameter in the credit parameter set includes: Determining a weighted sum of each of the credit parameters in the credit parameter set as a first target sum; Determining the sum of each of the credit parameters in the credit parameter set as a second target sum; The ratio of the first target sum to the second target sum is determined as the target credit value of the object to be identified.

5. The method according to claim 3, characterized in that: Determining multiple credit parameters of the to-be-identified object according to the credit values ​​of each of the target objects in the first object subset includes: Acquire an associated object of a first target object in the first object subset to obtain an associated object set of the first target object, wherein the first target object is any target object in the first object subset, the associated object is an object having the associated relationship with the first target object, and the associated object set includes the object to be identified; Determine the number of the associated objects in the associated object set to obtain a first quantity value; A first credit parameter of the object to be identified is determined by using the first quantity value and the credit value of the first target object, wherein the credit parameter set includes the first credit parameter.

6. The method according to claim 5, characterized in that Determining a first credit parameter of the to-be-identified object by using the first quantity value and the credit value of the first target object includes: Obtaining a ranking of the object to be identified in the associated object set to obtain a ranking value of the object to be identified, wherein each object in the associated object set is sorted according to the degree of association with the first target object; A first credit value of the object to be identified is determined according to the credit value of the first target object, the first quantity, and the ranking value.

7. The method according to claim 6, characterized in that Determining a first credit parameter of the to-be-identified object according to the credit value of the first target object, the first quantity, and the ranking value includes: Obtaining a target parameter value according to the first quantity and the ranking value; The product of the target parameter value and the credit value of the first target object is determined as the first credit parameter.

8. The method according to claim 1, characterized in that After determining whether the object to be identified is a candidate object according to the similarity, the method further includes: In the case that the object to be identified is the candidate object, a first identifier is assigned to the candidate object, wherein the target object in the first object set is assigned a second identifier.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program implements the steps of the method described in any one of claims 1 to 8 when executed by a processor.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to run the computer program to perform the method according to any one of claims 1 to 8.

Images