Customer risk prediction method, device, equipment and storage medium

Abstract

The present invention relates to the field of computer technology, and discloses a customer risk prediction method, device, equipment, and storage medium. The method includes: obtaining customer data, extracting triples of different ontologies based on the customer data, and establishing an ontology knowledge base based on the extraction results; using a Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and updating the ontology knowledge base based on the new triples; performing a first label labeling process on the triples before reasoning, and performing a second label labeling process on the new triples after reasoning, and associating the first label with the second label; constructing a knowledge graph based on the labeling process results and the associating process results, performing customer risk prediction based on the knowledge graph, and obtaining a customer risk prediction result. Through the above-mentioned method, the present invention can save computing resources, improve the efficiency and accuracy of customer risk prediction, and thus enhance user experience.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a customer risk prediction method, device, equipment and storage medium. Background Art

[0002] Nowadays, more and more companies are promoting digital transformation, and their data assets are increasing rapidly. How to reasonably use data assets to support and develop the company's main business has become the key to whether the company can stand out in the fierce market competition.

[0003] In all walks of life, customer risks, such as credit risk or financial risk, play an increasingly important role in daily life. Therefore, many industries have established models based on customer data collected or owned by the industry to predict customer risks.

[0004] At present, the industry's customer risk models are generally based on massive customer data. Key fields that affect customer risk are extracted from the data, and then weights are assigned based on the data values of the fields. Various classification and clustering algorithms are then used for machine learning training, ultimately resulting in a mature customer risk model for customer risk assessment and scoring.

[0005] This model has the following disadvantages when applied:

[0006] 1. Training models through machine learning consumes a lot of computing resources and time, and once customer data changes, resources need to be reinvested to update the model.

[0007] 2. Except for the algorithm implementers, others can only judge customer risks by scoring. The quantitative standards are single, and they do not know the specific judgment methods, making it impossible to effectively handle various specific situations.

[0008] 3. Machine learning models cannot provide relevant risk explanations or risk causes, but can only provide risk levels, which can easily lead to customer complaints and reports.

[0009] 4. The customer database is separated from business experts, and there is no visual and intuitive display of the relevant logic of customer risks. It is difficult for business experts to make effective parameter weight adjustments based on changes in customer data. Summary of the Invention

[0010] The present invention provides a customer risk prediction method, device, equipment and storage medium, which can save computing resources, improve the efficiency and accuracy of customer risk prediction, and thus enhance the user experience.

[0011] To solve the above technical problems, the present invention adopts a technical solution: providing a customer risk prediction method, comprising:

[0012] Acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results;

[0013] Using the Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and updating the ontology knowledge base according to the new triples;

[0014] Performing a first label tagging process on the triple before inference, performing a second label tagging process on the new triple after inference, and associating the first label with the second label;

[0015] A knowledge graph is constructed based on the labeling processing results and the association processing results, and customer risk prediction is performed based on the knowledge graph to obtain a customer risk prediction result.

[0016] According to one embodiment of the present invention, the triple includes an ontology class, an ontology instance, and an ontology attribute. The knowledge graph is constructed based on the annotation processing results and the association processing results. Customer risk prediction is performed based on the knowledge graph. The customer risk prediction result obtained includes:

[0017] Build a knowledge graph based on the annotation processing results and the association processing results;

[0018] Get the target ontology class input by the user;

[0019] Determine a first target label according to the target ontology class, and predict the cause of risk according to the first target label and the knowledge graph to obtain a first prediction result;

[0020] Determine an associated second target tag based on the first target tag, and predict the risk type and risk level based on the second target tag and the knowledge graph to obtain a second prediction result;

[0021] A customer risk prediction result of the target customer is determined based on the first prediction result and the second prediction result.

[0022] According to one embodiment of the present invention, after determining the customer risk prediction result of the target customer based on the first prediction result and the second prediction result, the method further includes:

[0023] generating a risk cause reasoning process record according to the first prediction result;

[0024] generating risk warning information according to the second prediction result;

[0025] The risk cause reasoning process record and the risk warning information are displayed on the front-end page.

[0026] According to one embodiment of the present invention, the step of constructing a knowledge graph based on the annotation processing results and the association processing results, performing customer risk prediction based on the knowledge graph, and obtaining the customer risk prediction result further includes:

[0027] Generate risk warning information based on the customer risk prediction result, and display the customer risk prediction result and the risk warning information on the front-end page.

[0028] According to one embodiment of the present invention, the using of the Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and updating the ontology knowledge base according to the new triples includes:

[0029] Obtaining preconfigured inference rules, loading the inference rules into the Jena inference engine, parsing the inference rules, and establishing an inference model;

[0030] Using the reasoning model to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology;

[0031] The ontology knowledge base is updated according to the new triples.

[0032] According to one embodiment of the present invention, the triples include an ontology class, an ontology instance, and an ontology attribute. The new triples corresponding to each ontology obtained by reasoning based on the ontology knowledge base using the reasoning model include:

[0033] Reading the first relationship between the ontology class, the ontology instance and the ontology attribute of each triple in the ontology knowledge base;

[0034] Using the reasoning model, through semantic reasoning, and based on the first relationship, a second relationship between a new ontology class, a new ontology instance, or a new ontology attribute is established;

[0035] The new triple is constructed according to the first relationship and the second relationship.

[0036] According to one embodiment of the present invention, after updating the ontology knowledge base according to the new triple, the method further includes:

[0037] performing iterative reasoning based on the updated ontology knowledge base;

[0038] The ontology knowledge base is updated according to the iterative reasoning result.

[0039] To solve the above technical problems, another technical solution adopted by the present invention is to provide a customer risk prediction device, comprising:

[0040] An acquisition module, configured to acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results;

[0041] A reasoning module, configured to use a Jena reasoning engine to perform reasoning based on the ontology knowledge base, obtain new triples corresponding to each ontology, and update the ontology knowledge base according to the new triples;

[0042] a labeling module, configured to label the triples before inference with a first label, label the new triples after inference with a second label, and associate the first label with the second label;

[0043] The construction and prediction module is used to construct a knowledge graph based on the annotation processing results and the association processing results, and perform customer risk prediction based on the knowledge graph to obtain customer risk prediction results.

[0044] In order to solve the above technical problems, another technical solution adopted by the present invention is: to provide a computer device, including: a memory, a processor and a computer program stored in the memory and runnable on the processor, and the processor implements the customer risk prediction method when executing the computer program.

[0045] In order to solve the above technical problems, another technical solution adopted by the present invention is: providing a computer storage medium on which a computer program is stored, and the computer program implements the above customer risk prediction method when executed by a processor.

[0046] The beneficial effects of the present invention are: reasoning is performed based on the ontology knowledge base through the Jena inference engine to obtain new triples corresponding to each ontology, the ontology knowledge base is updated according to the new triples, the triples before reasoning are annotated with the first label, the new triples after reasoning are annotated with the second label, the first label and the second label are associated, a knowledge graph is constructed according to the annotating processing results and the association processing results, customer risk prediction is performed based on the knowledge graph, and a customer risk prediction result is obtained. Compared with the machine learning training model, it can save computing resources, improve the efficiency of customer risk prediction, and can update the ontology knowledge base and knowledge graph in real time to improve the accuracy of customer risk prediction, thereby improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 1 is a flow chart of a customer risk prediction method according to an embodiment of the present invention;

[0048] Figure 2 1 is a flow chart of step S102 in the customer risk prediction method according to an embodiment of the present invention;

[0049] Figure 31 is a flow chart of step S104 in the customer risk prediction method according to an embodiment of the present invention;

[0050] Figure 4 is a flow chart of a customer risk prediction method according to another embodiment of the present invention;

[0051] Figure 5 is a flow chart of a customer risk prediction method according to another embodiment of the present invention;

[0052] Figure 6 2 is a schematic structural diagram of a customer risk prediction device according to an embodiment of the present invention;

[0053] Figure 7 is a schematic structural diagram of a computer device according to an embodiment of the present invention;

[0054] Figure 8 It is a schematic diagram of the structure of a computer storage medium according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0056] The terms "first", "second" and "third" in the present invention are used only for descriptive purposes and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, a feature defined as "first", "second" and "third" may explicitly or implicitly include at least one of such features. In the description of the present invention, "multiple" means at least two, for example, two, three, etc., unless otherwise clearly and specifically defined. All directional indications in the embodiments of the present invention (such as up, down, left, right, front, back...) are only used to explain the relative positional relationship, movement, etc. between the components under a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0057] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0058] Figure 1 It is a flow chart of a method for predicting customer risk according to an embodiment of the present invention. It should be noted that the method of the present invention is not limited to the method of predicting customer risk according to an embodiment of the present invention. Figure 1 The process sequence shown is limited. Figure 1 As shown, the method includes the steps of:

[0059] Step S101: Acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results.

[0060] In step S101, customer data is collected by users based on business needs. Triples consist of ontology classes, ontology instances, and ontology attributes. Ontology classes include risk, customer behavior, and customer information. Ontology instances include credit risk, bankruptcy liquidation, and defaulted loans. Ontology attributes include risky behavior and kinship relationships. This example uses Protégé to construct different ontology .owl files and establish an ontology knowledge base based on these different ontology .owl files.

[0061] Step S102: Use the Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and update the ontology knowledge base according to the new triples.

[0062] In step S102, the Jena inference engine is an important tool for implementing semantic logical reasoning of the ontology knowledge base. Developers use a large amount of historical customer data and experts' historical experience to write inference rules, and use the Jena inference engine to execute the inference rules based on the ontology knowledge base to obtain new triples of each ontology.

[0063] Further, see Figure 2 , step S102 further includes:

[0064] Step S201: Obtain pre-configured inference rules, load the inference rules into the Jena inference engine, parse the inference rules, and establish an inference model.

[0065] In step S201, the CreateDataset method of the TDBFactory class in the Jena inference engine is first used to create a new TDB object in the server's local directory. The ModelFactory class is then used to create an ontology model and read the ontology .owl file previously constructed in Protégé. The read ontology model is then added to the newly created TDB object. The constructed ontology library TDB database and the ontology model object are then read. A List collection is created to store the Jena inference engine's parsing results for the inference rules. The preconfigured inference rules in the Rules file are then read, converted into an input file stream, and read into the cache before being passed to the Jena inference engine for rule parsing. The parsed and loaded inference rules are then passed to the Jena inference engine, and the inference model is created and inference begins.

[0066] The inference rules of this embodiment are written in the Jena Rules standard language and are established according to the following general formula: [Rule name: (triplet combination 1), (triplet combination 2), …, (triplet combination N) -> (triplet combination 1), (triplet combination 2), …, (triplet combination N)]. The triple combination containing ontology classes, ontology instances, or ontology attributes is established as a condition on the left side of the arrow, and the triple combination of ontology classes, ontology instances, or ontology attributes corresponding to the conditions on the left side of the arrow is established on the right side of the arrow. The Jena inference engine infers the triple information corresponding to different ontologies in the ontology knowledge base based on the requirements on the left side of the arrow, and adds the triple information on the right side of the arrow to the ontologies that meet the rules.

[0067] Step S202: Use the reasoning model to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology.

[0068] In step S202, the first relationship between the ontology class, ontology instance and ontology attribute of each triple in the ontology knowledge base is read; the second relationship between the new ontology class, new ontology instance or new ontology attribute is established based on the first relationship by using the reasoning model through semantic reasoning; and the new triple is constructed according to the first relationship and the second relationship.

[0069] Specifically, any one of the ontology class, ontology instance and ontology attribute can be a variable. Assuming that the ontology class is a variable, the reasoning result is a new ontology class, and the new triple includes the new ontology class, the original ontology instance and the original ontology attribute; assuming that the ontology instance is a variable, the reasoning result is a new ontology instance, and the new triple includes the original ontology class, the new ontology instance and the original ontology attribute; assuming that the ontology attribute is a variable, the reasoning result is a new ontology attribute, and the new triple includes the original ontology class, the original ontology instance and the new ontology attribute.

[0070] Step S203: Update the ontology knowledge base according to the new triples.

[0071] In step S203 , the ontology knowledge base includes new triples after reasoning and old triples before reasoning.

[0072] Furthermore, in one achievable embodiment, step S203 further includes: performing iterative reasoning based on the updated ontology knowledge base; and updating the ontology knowledge base based on the iterative reasoning results. Specifically, the new triples in the updated ontology knowledge base are used as the old triples and the above reasoning process is repeated. Based on each reasoning result, the ontology knowledge base is updated and the reasoning steps are repeated. This embodiment improves reasoning accuracy through iterative reasoning and enables real-time updates of the ontology knowledge base, thereby ensuring the reliability of the reasoning data.

[0073] Step S103: perform a first labeling process on the triples before inference, perform a second labeling process on the new triples after inference, and associate the first label with the second label.

[0074] In step S103, the triples before inference are important data for subsequent risk cause analysis, and the new triples after inference are important data for subsequent risk type and risk level prediction. Therefore, these two types of data are labeled to improve data tracing efficiency, thereby improving customer risk prediction efficiency.

[0075] Step S104: construct a knowledge graph based on the labeling processing results and the association processing results, perform customer risk prediction based on the knowledge graph, and obtain a customer risk prediction result.

[0076] In step S104, the knowledge graph is used to represent the triples of different ontologies after annotation processing. The risk type, risk level and risk cause can be given through reasoning using the knowledge graph. The reasoning path can be traced throughout the process, thereby improving customer trust and reducing customer complaints.

[0077] Further, in one possible embodiment, see Figure 3 , step S104 includes:

[0078] Step S301: Construct a knowledge graph based on the annotation processing results and the association processing results.

[0079] Step S302: Obtain the target ontology class input by the user.

[0080] Specifically, the target ontology classes include risk, customer behavior, customer information, etc., for example, customer name, phone number, ID card information, etc.

[0081] Step S303: Determine a first target label according to the target ontology class, predict the risk cause according to the first target label and the knowledge graph, and obtain a first prediction result.

[0082] Specifically, the first target label is determined according to the target ontology class. According to the first target label, the position of the triple containing the first target label in the knowledge graph can be located. According to the triple, the risk cause is obtained. The first prediction result is the customer risk cause, such as credit record, bankruptcy liquidation, overdue mortgage, etc.

[0083] Step S304: Determine the associated second target tag based on the first target tag, and perform risk level prediction based on the second target tag and the knowledge graph to obtain a second prediction result.

[0084] Specifically, the second target label can be determined based on the first target label, and the position of the triple containing the second target label in the knowledge graph can be located based on the second target label. The risk level is obtained based on the triple. The second prediction result is the customer risk type and risk level. The risk type includes credit risk, financial risk, refusal to pay risk, insurance fraud risk, etc., and the risk level includes high risk, medium risk or low risk.

[0085] Step S305: Determine the customer risk prediction result of the target customer based on the first prediction result and the second prediction result.

[0086] Specifically, the customer risk prediction result may include the first prediction result and / or the second prediction result, and the customer risk prediction result may be displayed in the form of a graph or a report.

[0087] Further, in one possible embodiment, see Figure 4 After step S305, the method further includes:

[0088] Step S306: Generate a risk cause reasoning process record based on the first prediction result.

[0089] This embodiment can provide customers with relevant risk explanations through risk cause reasoning process records, avoid customer complaints and reports, and thus improve user experience.

[0090] Step S307: Generate risk warning information according to the second prediction result.

[0091] This embodiment can promptly remind users to judge customer risks through risk warning information to avoid unnecessary losses.

[0092] Step S308: Display the risk cause reasoning process record and risk warning information on the front-end page.

[0093] Specifically, the risk cause reasoning process records and risk warning information can be displayed in the form of graphs or reports.

[0094] The customer risk prediction method of one embodiment of the present invention uses the Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, updates the ontology knowledge base according to the new triples, performs first labeling processing on the triples before reasoning, performs second labeling processing on the new triples after reasoning, associates the first label with the second label, constructs a knowledge graph based on the labeling processing results and the association processing results, performs customer risk prediction based on the knowledge graph, and obtains customer risk prediction results. Compared with the machine learning training model, it can save computing resources and improve the efficiency of customer risk prediction. It can update the ontology knowledge base and knowledge graph in real time, improve the accuracy of customer risk prediction, and thus enhance the user experience.

[0095] Figure 5 It is a flow chart of a method for predicting customer risk according to another embodiment of the present invention. It should be noted that if there is substantially the same result, the method of the present invention is not limited to the method of Figure 5 The process sequence shown is limited. Figure 5 As shown, the method includes the steps of:

[0096] Step S501: Acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results.

[0097] In this embodiment, Figure 2 Step S201 and Figure 1 The process is similar to step S101 in , and for the sake of simplicity, it will not be repeated here.

[0098] Step S502: Use the Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and update the ontology knowledge base according to the new triples.

[0099] In this embodiment, Figure 5 Step S502 and Figure 1 The process is similar to step S102 in , and for the sake of simplicity, it will not be repeated here.

[0100] Step S503: perform a first labeling process on the triples before inference, perform a second labeling process on the new triples after inference, and associate the first label with the second label.

[0101] In this embodiment, Figure 5 Step S503 and Figure 1 The process is similar to step S103 in , and for the sake of brevity, it will not be repeated here.

[0102] Step S504: construct a knowledge graph based on the labeling processing results and the association processing results, perform customer risk prediction based on the knowledge graph, and obtain a customer risk prediction result.

[0103] In this embodiment, Figure 5 Step S504 and Figure 1 The process is similar to step S104 in the embodiment and will not be described again for the sake of brevity.

[0104] Step S505: Generate risk warning information based on the customer risk prediction results, and display the customer risk prediction results and risk warning information on the front-end page.

[0105] In step S505, the customer risk prediction results include the customer risk type, risk level, and customer risk cause. The customer risk prediction results and risk warning information are displayed in the form of a graph or report. This embodiment generates risk warning information based on the customer risk prediction results. This risk warning information can promptly remind users to assess customer risks and avoid unnecessary losses.

[0106] The customer risk prediction method of the embodiment of the present invention is based on the above embodiment and generates risk warning information according to the customer risk prediction results. The risk warning information can timely remind users to judge customer risks and avoid unnecessary losses.

[0107] Figure 6 Schematic diagram of the structure of the customer risk prediction device according to an embodiment of the present invention. Figure 6 As shown, the device 60 includes an acquisition module 61 , a reasoning module 62 , a labeling module 63 and a construction and prediction module 64 .

[0108] The acquisition module 61 is used to acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results;

[0109] The reasoning module 62 is used to use the Jena reasoning engine to perform reasoning based on the ontology knowledge base, obtain new triples corresponding to each ontology, and update the ontology knowledge base according to the new triples;

[0110] The labeling module 63 is used to label the triples before reasoning with a first label, label the new triples after reasoning with a second label, and associate the first label with the second label;

[0111] The construction and prediction module 64 is used to construct a knowledge graph based on the labeling processing results and the association processing results, perform customer risk prediction based on the knowledge graph, and obtain customer risk prediction results.

[0112] See also Figure 7 , Figure 7FIG. 1 is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. Figure 7 As shown, the computer device 70 includes a processor 71 and a memory 72 coupled to the processor 71 .

[0113] The memory 72 stores program instructions for implementing the customer risk prediction method described in any of the above embodiments.

[0114] The processor 71 is configured to execute program instructions stored in the memory 72 to predict customer risk.

[0115] The processor 71 may also be referred to as a CPU (Central Processing Unit). The processor 71 may be an integrated circuit chip having signal processing capabilities. The processor 71 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component. The general-purpose processor may be a microprocessor or any conventional processor.

[0116] See Figure 8 , Figure 8 Schematic diagram of the structure of the computer storage medium of the embodiment of the present invention. The computer storage medium of the embodiment of the present invention stores a program file 81 that can implement all the above methods, wherein the program file 81 can be stored in the above computer storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned computer storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or terminal devices such as a computer, a server, a mobile phone, and a tablet.

[0117] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0118] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0119] The above are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A customer risk prediction method, characterized in that: include: Acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results; Using the Jena inference engine to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and updating the ontology knowledge base according to the new triples; Performing a first label tagging process on the triple before inference, performing a second label tagging process on the new triple after inference, and associating the first label with the second label; Constructing a knowledge graph based on the labeling processing results and the association processing results, and performing customer risk prediction based on the knowledge graph to obtain a customer risk prediction result; The triples before reasoning include ontology classes, ontology instances, and ontology attributes. The knowledge graph is constructed based on the annotation processing results and the association processing results. Customer risk prediction is performed based on the knowledge graph. The customer risk prediction results obtained include: Build a knowledge graph based on the annotation processing results and the association processing results; Get the target ontology class input by the user; Determine a first target label according to the target ontology class, and predict the cause of risk according to the first target label and the knowledge graph to obtain a first prediction result; Determine an associated second target tag based on the first target tag, and predict the risk type and risk level based on the second target tag and the knowledge graph to obtain a second prediction result; A customer risk prediction result of the customer is determined based on the first prediction result and the second prediction result.

2. The customer risk prediction method according to claim 1, characterized in that: After determining the customer risk prediction result of the customer according to the first prediction result and the second prediction result, the method further includes: generating a risk cause reasoning process record according to the first prediction result; generating risk warning information according to the second prediction result; The risk cause reasoning process record and the risk warning information are displayed on the front-end page.

3. The customer risk prediction method according to claim 1, characterized in that: The method further includes: constructing a knowledge graph based on the labeling processing results and the association processing results, performing customer risk prediction based on the knowledge graph, and obtaining the customer risk prediction result; Generate risk warning information based on the customer risk prediction result, and display the customer risk prediction result and the risk warning information on the front-end page.

4. The customer risk prediction method according to claim 1, characterized in that: The Jena inference engine is used to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology, and updating the ontology knowledge base according to the new triples includes: Obtaining preconfigured inference rules, loading the inference rules into the Jena inference engine, parsing the inference rules, and establishing an inference model; Using the reasoning model to perform reasoning based on the ontology knowledge base to obtain new triples corresponding to each ontology; The ontology knowledge base is updated according to the new triples.

5. The customer risk prediction method according to claim 4, characterized in that: The triples before reasoning include ontology classes, ontology instances, and ontology attributes. The new triples corresponding to the ontology obtained by reasoning based on the ontology knowledge base using the reasoning model include: Reading the first relationship between the ontology class, the ontology instance and the ontology attribute of each triple in the ontology knowledge base; Using the reasoning model, through semantic reasoning, and based on the first relationship, a second relationship between a new ontology class, a new ontology instance, or a new ontology attribute is established; The new triple is created according to the first relationship and the second relationship.

6. The customer risk prediction method according to claim 4, characterized in that: After updating the ontology knowledge base according to the new triples, the method further includes: performing iterative reasoning based on the updated ontology knowledge base; The ontology knowledge base is updated according to the iterative reasoning result.

7. A customer risk prediction device, characterized in that: include: An acquisition module, configured to acquire customer data, extract triples of different ontologies based on the customer data, and establish an ontology knowledge base based on the extraction results; A reasoning module, configured to use a Jena reasoning engine to perform reasoning based on the ontology knowledge base, obtain new triples corresponding to each ontology, and update the ontology knowledge base according to the new triples; a labeling module, configured to label the triples before inference with a first label, label the new triples after inference with a second label, and associate the first label with the second label; A construction and prediction module is used to construct a knowledge graph based on the annotation processing results and the association processing results, and perform customer risk prediction based on the knowledge graph to obtain customer risk prediction results; The triples before reasoning include ontology classes, ontology instances, and ontology attributes. The knowledge graph is constructed based on the annotation processing results and the association processing results. Customer risk prediction is performed based on the knowledge graph. The customer risk prediction results obtained include: Build a knowledge graph based on the annotation processing results and the association processing results; Get the target ontology class input by the user; Determine a first target label according to the target ontology class, and predict the cause of risk according to the first target label and the knowledge graph to obtain a first prediction result; Determine an associated second target tag based on the first target tag, and predict the risk type and risk level based on the second target tag and the knowledge graph to obtain a second prediction result; A customer risk prediction result of the customer is determined based on the first prediction result and the second prediction result.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the customer risk prediction method according to any one of claims 1 to 6 when executing the computer program.

9. A computer storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the customer risk prediction method according to any one of claims 1 to 6 is implemented.

Images