Personal credit assessment method and device, equipment and storage medium
By introducing the NSGA-II genetic algorithm in personal credit assessment, a credit evaluation model with multiple evaluation indicators is constructed, which solves the problem that traditional methods are difficult to accurately evaluate personal credit, and achieves more efficient and accurate credit evaluation.
Patent Information
- Application Number
- CN202411726779.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-28
- Publication Date
- 2025-02-18
AI Technical Summary
Traditional personal credit assessment methods are difficult to accurately reflect the nonlinearity and uncertainty of the market, and it is difficult to take into account the influence of multiple factors, resulting in inaccurate assessment and low prevention and control efficiency.
The NSGA-II genetic algorithm is used to build a credit evaluation model containing multiple evaluation indicators. By obtaining user basic information, asset status and personal expectations, a model of primary evaluation indicators and secondary evaluation indicators is established, and a genetic algorithm is used to perform the Calculate to obtain personal credit evaluation results.
It improves the accuracy and comprehensiveness of credit evaluation, enhances the persuasiveness of the evaluation results, and continuously approaches the optimal solution through iterative calculations, providing more accurate credit evaluation results.
Smart Images
Figure CN119477524A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of financial technology, and in particular to a personal credit assessment method, device, equipment and storage medium. Background Art
[0002] While credit business brings benefits, it also comes with certain risks. Bank credit risk control is of great significance in ensuring the safety of bank assets, promoting sustainable business development, maintaining financial market stability, protecting consumer rights and interests, and complying with regulatory requirements. In the credit business, the core issue to be focused on is how to accurately assess personal credit. Personal credit assessment is to analyze the individual's indicator data information through certain methods based on the basic personal situation and related environmental factors to determine the individual's credit rating. Traditional risk assessment methods often rely on historical data and simple statistical models. However, these methods often ignore the nonlinearity and uncertainty of the market, lack global considerations, and are difficult to take into account the influence of multiple factors. Sometimes, inaccurate assessments and low prevention and control efficiency may occur.
[0003] Therefore, obtaining a more accurate, comprehensive and efficient personal credit assessment method to cope with the complexity and uncertainty in credit business is crucial to improving the bank's credit risk control capabilities. Summary of the invention
[0004] In response to the above problems, this application provides a personal credit assessment method, including the following contents:
[0005] In a first aspect, the present application provides a method for evaluating personal credit, the method comprising:
[0006] Obtaining user information, which includes at least basic information, asset status and personal expectations;
[0007] A credit assessment model including multiple evaluation indicators is established based on the user information, wherein the indicators include primary evaluation indicators and secondary evaluation indicators; the secondary evaluation indicators are specific subdivision indicators of the primary evaluation indicators, and are used to further explain and illustrate the content of the primary evaluation indicators;
[0008] Based on the credit evaluation model, the NSGA-II genetic algorithm is used to calculate the user information to obtain the personal credit evaluation result.
[0009] Optionally, the credit assessment model is:
[0010]
[0011] Among them, y is the value of personal credit; w i is the allocation weight of the i-th primary evaluation index, wijk is the weight assigned to the kth attribute of the ijth indicator; the ijth indicator represents the jth secondary indicator under the ith first-level evaluation indicator; w i and w ijk is the gene in the genetic algorithm, m represents a total of m first-level indicators, n represents that the ith first-level indicator has n second-level indicators; l represents that the ijth indicator has l possible values; φijk(x ij ,y ijk ) is a 0-1 function, expressed as follows:
[0012]
[0013] x represents the allocation weight of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator, and y represents the value of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator.
[0014] Optionally, after obtaining the credit assessment model, the method further includes:
[0015] Using binary coding, each indicator and its attributes are used as genes in the genetic algorithm for encoding operations to obtain the chromosomes in the genetic algorithm.
[0016] Optionally, the method further includes:
[0017] A fitness function is established, and the judgment accuracy for customers with good credit and the judgment accuracy for customers with poor credit are used as individual fitness evaluations. By comparing the fitness, individuals with higher fitness are selected to survive and their offspring have better adaptability through genetic mechanisms. The fitness function is:
[0018]
[0019] Among them, fit g Indicates the accuracy of judgment for customers with good credit, fit b Indicates the accuracy of judgment on customers with poor credit; n g and n b Represents the number of customers with good credit and customers with poor credit, m g and m b Represent the number of misjudgments among customers with good credit and customers with poor credit, respectively.
[0020] Optionally, the method further includes:
[0021] Determining a genetic operator, wherein the genetic operator is used to generate new individuals to ensure the continuous evolution of the population;
[0022] The determination of the genetic operator specifically includes designing a selection operator, a crossover operator and a mutation operator.
[0023] In a second aspect, the present application provides a personal credit assessment device, the device comprising:
[0024] An acquisition unit, used to acquire user information, wherein the user information at least includes basic information, asset status and personal expectations;
[0025] An establishing unit, used to establish a credit assessment model including a plurality of evaluation indicators based on the user information, wherein the indicators include a primary evaluation indicator and a secondary evaluation indicator; the secondary evaluation indicator is a specific subdivision indicator of the primary evaluation indicator, and is used to further explain and illustrate the content of the primary evaluation indicator;
[0026] The calculation unit is used to calculate the user information based on the credit evaluation model using the NSGA-II genetic algorithm to obtain a personal credit evaluation result.
[0027] Optionally, the credit assessment model is:
[0028]
[0029] Among them, y is the value of personal credit; w i is the allocation weight of the i-th primary evaluation index, w ijk is the weight assigned to the kth attribute of the ijth indicator; the ijth indicator represents the jth secondary indicator under the ith first-level evaluation indicator; w i and w ijk is the gene in the genetic algorithm, m represents a total of m first-level indicators, n represents that the ith first-level indicator has n second-level indicators; l represents that the ijth indicator has l possible values; φijk(x ij ,y ijk ) is a 0-1 function, expressed as follows:
[0030]
[0031] x represents the allocation weight of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator, and y represents the value of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator.
[0032] Optionally, the computing unit is further used for:
[0033] Using binary coding, each indicator and its attributes are used as genes in the genetic algorithm for encoding operations to obtain the chromosomes in the genetic algorithm.
[0034] Optionally, the establishing unit is further used for:
[0035] A fitness function is established, and the judgment accuracy for customers with good credit and the judgment accuracy for customers with poor credit are used as individual fitness evaluations. By comparing the fitness, individuals with higher fitness are selected to survive and their offspring have better adaptability through genetic mechanisms. The fitness function is:
[0036]
[0037] Among them, fit g Indicates the accuracy of judgment for customers with good credit, fit b Indicates the accuracy of judgment on customers with poor credit; n g and n b Represents the number of customers with good credit and customers with poor credit, m g and m b Represent the number of misjudgments among customers with good credit and customers with poor credit, respectively.
[0038] Optionally, the computing unit is further used for:
[0039] Determining a genetic operator, wherein the genetic operator is used to generate new individuals to ensure the continuous evolution of the population;
[0040] The determination of the genetic operator specifically includes designing a selection operator, a crossover operator and a mutation operator.
[0041] In a third aspect, the present application provides a device, comprising a memory and a processor, wherein the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device performs the personal credit assessment method introduced in any implementation of the first aspect.
[0042] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a code. When the code is executed, a device executing the code implements the personal credit assessment method introduced in any implementation of the first aspect.
[0043] This application provides a personal credit assessment method. By obtaining diversified information such as the user's basic information, asset status, and personal expectations, the user's credit status can be fully and accurately reflected. Furthermore, a credit assessment model based on this information that includes multiple evaluation indicators (including primary evaluation indicators and secondary evaluation indicators) not only improves the accuracy of the credit assessment, but also makes the evaluation results more convincing. More importantly, this method innovatively introduces the NSGA-II genetic algorithm to calculate user information to obtain personal credit assessment results. As an advanced optimization algorithm, the NSGA-II genetic algorithm can efficiently handle complex multi-objective optimization problems. Through iterative calculations, the algorithm can continuously approach the optimal solution, thereby providing users with more accurate credit assessment results. BRIEF DESCRIPTION OF THE DRAWINGS
[0044] In order to more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0045] Figure 1 A flowchart of a personal credit assessment method provided in an embodiment of the present application;
[0046] Figure 2 A schematic diagram of a genetic algorithm flow chart provided in an embodiment of the present application;
[0047] Figure 3 A schematic diagram of the structure of a personal credit assessment device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0049] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.
[0050] Explanation of relevant terms
[0051] Genetic algorithm: Genetic algorithm is an optimization algorithm that simulates the biological evolution process. In genetic algorithm, the solution to the problem is encoded as the gene sequence of an individual, just like the genes of an organism. Each gene represents a component of a possible solution, and the collection of gene sequences forms a population. Its core is to optimize the solution by simulating the three biological evolution processes of selection, crossover, and mutation. First, based on the fitness evaluation of each individual, individuals with higher fitness are selected as parents of the next generation. This is similar to the mechanism in nature where organisms with strong adaptability are more likely to reproduce offspring. Then, the selected parent individuals are crossovered to generate new individuals through the combination and exchange of genes. This is similar to the combination and inheritance of genes when organisms reproduce. Through the crossover operation, the new individuals inherit the favorable characteristics of the parent individuals. Finally, the newly generated individuals are mutated to introduce random genetic changes. This is similar to the gene mutation in the process of biological evolution. The mutation operation increases the diversity of the population and helps to explore a wider solution space. By repeatedly performing selection, crossover, and mutation operations, the genetic algorithm gradually optimizes the gene sequence of individuals to find the optimal solution. This optimal solution is the solution with the best performance for a given problem.
[0052] NSGA-II algorithm: NSGA-II algorithm (Non-dominated Sorting Genetic Algorithm II) is a classic genetic algorithm for solving multi-objective optimization problems. It uses a fast non-dominated sorting algorithm, and the computational complexity is greatly reduced compared to NSGA; it uses crowding and crowding comparison operators to replace the shared radius that needs to be specified, and uses them as the winning criteria in the same-level comparison after fast sorting, so that individuals in the quasi-Pareto domain can be expanded to the entire Pareto domain and evenly distributed, maintaining the diversity of the population; it introduces an elite strategy, expands the sampling space, prevents the loss of the best individuals, and improves the algorithm's computing speed and robustness.
[0053] Non-dominated solution: also known as Pareto solution or undominated solution, is an important concept in multi-objective optimization problems. In multi-objective optimization problems, due to the possible conflicts and difficulties in comparison between various objectives, a solution may perform optimally on a certain objective and perform poorly on other objectives. Therefore, the concept of non-dominated solution came into being. It is defined as follows: Suppose there are two solutions S1 and S2. If S1 is better than S2 in all objectives, then S1 is said to dominate S2. On the contrary, if a solution S1 is not dominated by any other solution, then S1 is called a non-dominated solution. These non-dominated solutions constitute the so-called Pareto frontier. The solutions on this frontier are all non-dominated, which means that their performance on the objective function is optimal.
[0054] Figure 1 A flowchart of a personal credit assessment method provided in an embodiment of the present application. Figure 1 As shown, the personal credit assessment method provided in the embodiment of the present application may include:
[0055] S101. Obtain user information, which includes at least basic information, asset status and personal expectations.
[0056] The basic information includes gender, age, marital status, occupation, etc. The asset status includes income, liabilities, real estate, vehicles, etc. The personal expectations include future income expectations, loan purpose, etc.
[0057] The acquired user information is preprocessed, including data cleaning, missing value processing, outlier detection, etc., to ensure the integrity and accuracy of the information.
[0058] S102: establishing a credit assessment model including a plurality of evaluation indicators based on the user information, wherein the indicators include primary evaluation indicators and secondary evaluation indicators.
[0059] According to the characteristics of user information and business needs, determine the primary and secondary indicators of the evaluation model. The primary evaluation indicators are a large category in the credit evaluation model, which generally reflect different aspects of user credit evaluation. The primary indicators can include personal basic information, asset status, personal expectations, etc., while the secondary evaluation indicators are specific subdivisions of the primary evaluation indicators. They provide more detailed information to further explain and illustrate the content of the primary evaluation indicators. Such as gender, age, etc. under personal basic information. By building a credit evaluation model that includes primary and secondary evaluation indicators, you can have a more comprehensive understanding of the user's credit status and make a more accurate credit evaluation based on this information. At the same time, this model design also helps to flexibly adjust and optimize according to business needs.
[0060] In one implementation of the embodiment of the present application, the credit assessment model is:
[0061]
[0062] Among them, y is the value of personal credit; w i is the allocation weight of the i-th primary evaluation index, w ijk is the weight assigned to the kth attribute of the ijth indicator; the ijth indicator represents the jth secondary indicator under the ith first-level evaluation indicator; w i and w ijk is the gene in the genetic algorithm, m represents a total of m first-level indicators, n represents that the ith first-level indicator has n second-level indicators; l represents that the ijth indicator has l possible values; φijk(x ij ,y ijk ) is a 0-1 function, expressed as follows:
[0063]
[0064] x represents the allocation weight of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator, and y represents the value of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator.
[0065] S103: Based on the credit assessment model, the NSGA-II genetic algorithm is used to calculate the user information to obtain a personal credit assessment result.
[0066] NSGA-II is an advanced evolutionary algorithm that iteratively optimizes potential solutions (i.e., individuals) by simulating the process of natural selection to find a near-optimal solution set. The core mechanisms of NSGA-II include non-dominated sorting, crowding comparison, and elite strategy, which together ensure the effectiveness and efficiency of the algorithm.
[0067] In NSGA-II, each individual represents a potential solution. For the credit assessment model, individuals can be encoded as a set of all first-level evaluation index weights (w i ) and secondary evaluation index attribute weight (w ijk These weights together constitute the parameter set of the credit assessment model.
[0068] In an implementation of the embodiment of the present application, after obtaining the credit assessment model, the method further includes: using a binary encoding method to encode each indicator and its attributes as a gene in a genetic algorithm to obtain a chromosome in the genetic algorithm.
[0069] In the operation of genetic algorithm, it does not directly operate on the actual decision variables of the problem to be solved, but selects, crosses, mutates, and other operations on the individual codes representing feasible solutions. The purpose of optimization is achieved through this genetic operation, which is one of the characteristics of genetic algorithm. There are two commonly used encoding methods: binary encoding and floating point encoding. Binary encoding is the main encoding method in genetic algorithm. Since the credit assessment model constructed in this application has to determine two parameters, the number is small, and the value range is small, therefore, binary encoding is used, where w in the mathematical model i and w ijk This is the gene that operates this inheritance. Each individual will contain all w i and w ijk The value of directly constitutes a chromosome.
[0070] In one implementation of the embodiment of the present application, the method further includes:
[0071] Establish a fitness function, which is the key to measuring the quality of an individual. In the credit assessment scenario, the fitness function can be defined as a function of the model's prediction accuracy, stability, or other relevant performance indicators. Specifically, a training dataset containing users with known credit ratings can be constructed and used to evaluate the prediction accuracy of each individual. The higher the prediction accuracy, the better the individual's fitness.
[0072] The judgment accuracy for customers with good credit and the judgment accuracy for customers with poor credit are used as the fitness evaluation of individuals. By comparing the fitness, individuals with higher fitness are selected to survive and their offspring have better adaptability through genetic mechanisms. The fitness function is:
[0073]
[0074] Among them, fit g Indicates the accuracy of judgment for customers with good credit, fit b Indicates the accuracy of judgment on customers with poor credit; n g and n b Represents the number of customers with good credit and customers with poor credit, m g and m b Represent the number of misjudgments among customers with good credit and customers with poor credit, respectively.
[0075] In one implementation of the embodiment of the present application, the method further includes:
[0076] Determine a genetic operator, wherein the genetic operator is used to generate new individuals to ensure the continuous evolution of the population; the determined genetic operator specifically includes designing a selection operator, a crossover operator, and a mutation operator.
[0077] In genetic algorithms, genetic operators are mainly used to generate new individuals to ensure the continuous evolution of the population. For multi-objective optimization problems with fitness functions, this application uses the NSGA-II algorithm to solve them.
[0078] First, about the selection operator. The selection operator is based on the evaluation of the fitness of individuals. Its main purpose is to avoid gene loss and improve global convergence and running speed. NSGA-II uses a binary tournament as the selection operation. In each generation, two individuals are first randomly selected from the current population using the binary tournament selection operation, and then the individual with a higher non-dominated level is selected. This selection process is performed by comparing the non-dominated level and crowding distance of the individual.
[0079] Secondly, a crossover operator is designed. The crossover operator refers to exchanging some genes of two paired chromosomes in a certain way to form two new individuals. The present invention adopts single-point crossover, that is, a crossover point is randomly set in a specified individual, and then the genes after the point are exchanged. It is the most commonly used and basic crossover operator.
[0080] Finally, a mutation operator is designed. The principle of the mutation operator is similar to gene mutation, which is also a method of generating new individuals. This application adopts a basic bit mutation operator, that is, a mutation operation is performed on the value of a certain bit or several genes randomly specified with a mutation probability in the individual code.
[0081] In addition, NSGA-II adopts an elite strategy to retain the best individuals in each generation. This means that after each generation, the algorithm will directly copy the best individuals in the current generation to the next generation to ensure the continuity of excellent solutions. The algorithm will set a termination condition, such as reaching a preset number of iterations, finding a satisfactory solution, or the fitness is no longer significantly improved. When the termination condition is met, the algorithm will stop iterating and output the optimal solution set.
[0082] After iterative optimization by the NSGA-II genetic algorithm, a set of optimal weight sets (wi and wijk) can be obtained. These weight sets represent the optimal parameter configuration of the credit assessment model. Next, these weight sets can be substituted into the credit assessment model, and the user's actual information can be used to calculate the personal credit assessment results. Specifically, the model will calculate the score of each primary evaluation indicator based on the user's secondary evaluation indicator value and the corresponding weight, and finally weighted sum these scores to obtain the personal credit value (y). By combining the credit assessment model and the NSGA-II genetic algorithm, user information can be effectively calculated and personal credit assessment results can be obtained. This method not only improves the accuracy of credit assessment, but also enhances the flexibility and adaptability of the model.
[0083] The above embodiment introduces the personal credit assessment method in this application. Figure 2 A detailed introduction to the genetic algorithm in this application is given. Figure 2 The flowchart of a genetic algorithm provided in the embodiment of the present application describes in detail the entire process of the genetic algorithm from initialization to finally finding the optimal solution (or ending after reaching the maximum number of iterations). The following are the specific steps:
[0084] S201. Start.
[0085] S202: Initialize the population.
[0086] In a genetic algorithm, an initial population of multiple individuals is first generated. These individuals are generated randomly or obtained through some heuristic method. Each individual represents a possible solution, and its genetic code represents the specific characteristics of the solution.
[0087] S203: Determine whether to generate the first generation sub-population.
[0088] If yes, proceed to step S206; if no, proceed to step S204.
[0089] S204: Non-dominated sorting.
[0090] After the sub-population is generated, the new population after the merger of the parent and the offspring needs to be non-dominated sorted first. This step is to classify the individuals in the population according to their fitness (i.e. the performance of the personal credit assessment model) for subsequent selection operations. Non-dominated sorting is a commonly used sorting method in genetic algorithms, which compares the superiority and inferiority relationship between individuals based on their fitness values.
[0091] S205, selection, crossover, mutation.
[0092] Selection: Based on the results of non-dominated sorting, select individuals with higher fitness as the parents of the next generation. Here, strategies such as binary tournament selection can be used to ensure that the selected individuals have higher fitness and diversity.
[0093] Crossover: Perform a crossover operation on the selected parent individuals to generate new individuals (i.e. offspring) by combining and exchanging genes. The crossover operation can introduce new gene combinations, thereby increasing the diversity of the population.
[0094] Mutation: Perform mutation operations on newly generated individuals to randomly change the values of some of their genes. Mutation operations can further increase the diversity of the population and help explore a wider solution space.
[0095] These three steps are the core operations in the genetic algorithm. The selection operation is used to select individuals with higher fitness from the current population as parents; the crossover operation generates new offspring individuals by exchanging some genes of the parent individuals; and the mutation operation randomly perturbs the genes of the offspring individuals to increase the diversity of the population.
[0096] S206, evolutionary algebra Gen=2.
[0097] After completing the generation and genetic operation of the first generation of sub-populations, the evolutionary generation number Gen is set to 2, ready to enter the next round of iteration. Here, the current evolutionary generation number is 2, which means that one iteration has been performed. In the genetic algorithm, each generation will generate a new population through operations such as selection, crossover, and mutation.
[0098] S207, the parent and child individuals are merged.
[0099] Before each generation of evolution begins, the parent and offspring individuals are merged into a new population for the next generation of evolution. This is done to maintain the diversity and evolutionary potential of the population. The parent and offspring individuals are merged into a new population for the next generation of evolution.
[0100] S208: Determine whether to generate a new parent population.
[0101] A new parent population is directly generated by a selection operation, rather than by merging and then generating. If yes, then go to step S212, if no, then go to step S209.
[0102] S209, fast non-dominated sort.
[0103] Perform fast non-dominated sorting on the merged population to quickly determine the merits and demerits of individuals.
[0104] S210: Calculate the congestion degree.
[0105] In genetic algorithms, crowding is an important indicator to measure the relative position of individuals in the population. It is used to maintain the diversity of the population and prevent the algorithm from converging to a local optimal solution too early.
[0106] S211. Select suitable individuals to form a new parent population.
[0107] After reaching the maximum number of generations, the best individual (or combination) is selected from the current population to form a new parent population as the final output of the algorithm. The "combination" here may be a metaphor, but it actually refers to the best individual selected.
[0108] S212, selection, crossover, and mutation.
[0109] Perform genetic operations on the new parent population to generate a new child population.
[0110] S213. Determine whether Gen is less than the maximum algebra.
[0111] Check whether the current evolutionary generation has reached the preset maximum generation. If not, proceed to step S214 and continue to the next round of iteration; if reached, end the entire process.
[0112] S214, Gen=Gen+1.
[0113] Increase the evolutionary generation number Gen by 1 and prepare to enter the next round of iteration.
[0114] S215, end.
[0115] The above embodiment of the present application provides a personal credit evaluation method. By using a genetic algorithm to construct a personal credit evaluation model, it can not only improve the accuracy of the evaluation, but also effectively improve the optimization efficiency and search ability, and can better cope with market fluctuations; through the NSGA-II multi-objective optimization algorithm, a set of Pareto optimal solutions can be effectively retained, avoiding the determination of single-objective parameter weights, and can include all preferences and contain more effective information. In addition, the NSGA-II genetic algorithm used in this invention converts personal credit evaluation into a multi-objective optimization problem. The algorithm initializes the population and performs non-dominated sorting, selection, crossover, mutation and other steps to continuously evolve and generate new sub-populations and parent populations until the termination condition is met. In this process, the algorithm not only considers multiple evaluation indicators, including primary evaluation indicators and secondary evaluation indicators that are further subdivided and explained, but also realizes a comprehensive and efficient solution to the personal credit evaluation problem through the unique mechanism of the NSGA-II genetic algorithm, such as parent and child individual merging, fast non-dominated sorting and crowding calculation. This method not only significantly improves the solution efficiency and can obtain personal credit assessment results in a shorter time, but also fully meets various requirements, such as accuracy, stability and reliability. The NSGA-II genetic algorithm shows strong global search capabilities and robustness, which can effectively avoid falling into local optimal solutions, thereby more accurately assessing personal credit status. In addition, the algorithm also effectively overcomes the limitations of traditional methods, such as single assessment results and susceptibility to subjective factors, providing a more scientific, objective and efficient method for personal credit assessment.
[0116] The above are some specific implementations of a personal credit assessment method provided in the embodiment of the present application. Based on this, the present application also provides a corresponding device. The device provided in the embodiment of the present application will be introduced from the perspective of functional modularization.
[0117] Figure 3 A schematic diagram of the structure of a personal credit assessment device provided in an embodiment of the present application. Figure 3As shown, the personal credit assessment device 300 provided in the embodiment of the present application includes:
[0118] An acquisition unit 310 is used to acquire user information, wherein the user information at least includes basic information, asset status and personal expectations;
[0119] An establishing unit 320 is used to establish a credit assessment model including multiple evaluation indicators based on the user information, wherein the indicators include primary evaluation indicators and secondary evaluation indicators; the secondary evaluation indicators are specific subdivision indicators of the primary evaluation indicators, and are used to further explain and illustrate the content of the primary evaluation indicators;
[0120] The calculation unit 330 is used to calculate the user information based on the credit evaluation model using the NSGA-II genetic algorithm to obtain a personal credit evaluation result.
[0121] In one implementation of the embodiment of the present application, the credit assessment model is:
[0122]
[0123] Among them, y is the value of personal credit; w i is the allocation weight of the i-th primary evaluation index, w ijk is the weight assigned to the kth attribute of the ijth indicator; the ijth indicator represents the jth secondary indicator under the ith first-level evaluation indicator; w i and w ijk is the gene in the genetic algorithm, m represents a total of m first-level indicators, n represents that the ith first-level indicator has n second-level indicators; l represents that the ijth indicator has l possible values; φijk(x ij ,y ijk ) is a 0-1 function, expressed as follows:
[0124]
[0125] x represents the allocation weight of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator, and y represents the value of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator.
[0126] In one implementation of the embodiment of the present application, the computing unit is further configured to:
[0127] Using binary coding, each indicator and its attributes are used as genes in the genetic algorithm for encoding operations to obtain the chromosomes in the genetic algorithm.
[0128] In one implementation of the embodiment of the present application, the establishing unit is further configured to:
[0129] A fitness function is established, and the judgment accuracy for customers with good credit and the judgment accuracy for customers with poor credit are used as individual fitness evaluations. By comparing the fitness, individuals with higher fitness are selected to survive and their offspring have better adaptability through genetic mechanisms. The fitness function is:
[0130]
[0131] Among them, fit g Indicates the accuracy of judgment for customers with good credit, fit b Indicates the accuracy of judgment on customers with poor credit; n g and n b Represents the number of customers with good credit and customers with poor credit, m g and m b Represent the number of misjudgments among customers with good credit and customers with poor credit, respectively.
[0132] In one implementation of the embodiment of the present application, the computing unit is further configured to:
[0133] Determining a genetic operator, wherein the genetic operator is used to generate new individuals to ensure the continuous evolution of the population;
[0134] The determination of the genetic operator specifically includes designing a selection operator, a crossover operator and a mutation operator.
[0135] The embodiments of the present application also provide corresponding devices and computer storage media for implementing the solutions provided by the embodiments of the present application.
[0136] The device includes a memory and a processor, the memory is used to store instructions or codes, and the processor is used to execute the instructions or codes so that the device executes the method described in any embodiment of the present application.
[0137] The computer storage medium stores codes, and when the codes are executed, a device executing the codes implements the method described in any embodiment of the present application.
[0138] Through the description of the above implementation methods, it can be known that those skilled in the art can clearly understand that all or part of the steps in the above-mentioned embodiment method can be implemented by means of software plus a general hardware platform. Based on such an understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a storage medium, such as a read-only memory (ROM) / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network communication device such as a router) to execute the methods described in each embodiment of the present application or some parts of the embodiments.
[0139] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0140] It should also be noted that the various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device and apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments. The device and apparatus embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components indicated as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0141] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
Claims
1. A personal credit assessment method, characterized in that: The method comprises: Obtaining user information, which includes at least basic information, asset status and personal expectations; A credit assessment model including multiple evaluation indicators is established based on the user information, wherein the indicators include primary evaluation indicators and secondary evaluation indicators; the secondary evaluation indicators are specific subdivision indicators of the primary evaluation indicators, and are used to further explain and illustrate the content of the primary evaluation indicators; Based on the credit evaluation model, the NSGA-II genetic algorithm is used to calculate the user information to obtain the personal credit evaluation result.
2. The method according to claim 1, characterized in that The credit assessment model is: Among them, y is the value of personal credit; w i is the allocation weight of the i-th primary evaluation index, w ijk is the weight assigned to the kth attribute of the ijth indicator; the ijth indicator represents the jth secondary indicator under the ith first-level evaluation indicator; w i and w ijk is the gene in the genetic algorithm, m represents a total of m first-level indicators, n represents that the ith first-level indicator has n second-level indicators; l represents that the ijth indicator has l possible values; φijk(x ij ,y ijk ) is a 0-1 function, expressed as follows: x represents the allocation weight of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator, and y represents the value of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator.
3. The method according to claim 2, characterized in that After obtaining the credit assessment model, the method further includes: Using binary coding, each indicator and its attributes are used as genes in the genetic algorithm for encoding operations to obtain the chromosomes in the genetic algorithm.
4. The method according to claim 3, characterized in that: The method further comprises: A fitness function is established, and the judgment accuracy for customers with good credit and the judgment accuracy for customers with poor credit are used as individual fitness evaluations. By comparing the fitness, individuals with higher fitness are selected to survive and their offspring have better adaptability through genetic mechanisms. The fitness function is: Among them, fit g Indicates the accuracy of judgment for customers with good credit, fit b Indicates the accuracy of judgment on customers with poor credit; n g and n b Represents the number of customers with good credit and customers with poor credit, m g and m b Represent the number of misjudgments among customers with good credit and customers with poor credit, respectively.
5. The method according to claim 4, characterized in that The method further comprises: Determining a genetic operator, wherein the genetic operator is used to generate new individuals to ensure the continuous evolution of the population; The determination of the genetic operator specifically includes designing a selection operator, a crossover operator and a mutation operator.
6. A personal credit assessment device, characterized in that: The device comprises: An acquisition unit, used to acquire user information, wherein the user information at least includes basic information, asset status and personal expectations; An establishing unit, used to establish a credit assessment model including a plurality of evaluation indicators based on the user information, wherein the indicators include a primary evaluation indicator and a secondary evaluation indicator; the secondary evaluation indicator is a specific subdivision indicator of the primary evaluation indicator, and is used to further explain and illustrate the content of the primary evaluation indicator; The calculation unit is used to calculate the user information based on the credit evaluation model using the NSGA-II genetic algorithm to obtain a personal credit evaluation result.
7. The system according to claim 6, characterized in that The credit assessment model is: Among them, y is the value of personal credit; w i is the allocation weight of the i-th primary evaluation index, w ijk is the weight assigned to the kth attribute of the ijth indicator; the ijth indicator represents the jth secondary indicator under the ith first-level evaluation indicator; w i and w ijk is the gene in the genetic algorithm, m represents a total of m first-level indicators, n represents that the ith first-level indicator has n second-level indicators; l represents that the ijth indicator has l possible values; φijk(x ij ,y ijk ) is a 0-1 function, expressed as follows: x represents the allocation weight of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator, and y represents the value of the kth attribute of the jth secondary indicator under the ith primary evaluation indicator.
8. The system according to claim 7, characterized in that The computing unit is also used for: Using binary coding, each indicator and its attributes are used as genes in the genetic algorithm for encoding operations to obtain the chromosomes in the genetic algorithm.
9. A computing device, characterized in that The computing device includes: a memory and a processor; The memory is used to store computer programs; The processor is configured to implement the method according to any one of claims 1 to 5 when executing the computer program.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.