Insurance policy quality inspection method and device, computer equipment and storage medium

Abstract

The embodiment of the invention belongs to the field of financial science and technology, and relates to an insurance policy quality inspection method and device, computer equipment and a storage medium, and the method comprises the following steps: obtaining insurance policy data, and extracting key insurance policy basic information features, insured person features and insurance type features from the insurance policy data; constructing a risk assessment model based on the basic information features of the insurance policy, the insured person features and the insurance type features; performing insurance policy risk assessment on the insurance policy data based on an ant colony algorithm and the risk assessment model to obtain a risk distribution map; extracting high-risk insurance policies from the risk distribution map, and performing clustering analysis on the high-risk insurance policies to obtain risk analysis information; and generating an insurance policy quality inspection report based on the risk analysis information. According to the method, accurate and effective risk assessment and analysis can be effectively performed on the insurance policy, and the corresponding quality inspection report is generated and displayed.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, specifically to the field of financial technology, and in particular to a policy quality inspection method, device, computer equipment and storage medium. Background Art

[0002] In the insurance business system, the insurance policy, as the core component of the insurance business, carries key information such as insurance liability, policyholder information, insured information, insurance period, premium amount, etc. The quality of the insurance policy is directly related to the insurance company's operating risk, customer satisfaction, and business sustainability. Therefore, policy quality inspection has become an indispensable part of the insurance business. Its goal is to accurately identify high-risk abnormal policies in massive policy data, and provide an effective and reliable decision-making basis for subsequent risk analysis and resolution.

[0003] Traditional policy quality inspection methods mainly rely on manual work, which involves manually reviewing policy information to identify potential risk points. However, with the rapid development of insurance business and the sharp increase in the number of policies, manual quality inspection methods have gradually exposed problems such as low efficiency and serious waste of human resources. Manual quality inspection is not only time-consuming and labor-intensive, but also limited by personal experience and knowledge level, making it difficult to ensure the consistency and accuracy of quality inspection.

[0004] In order to overcome the limitations of traditional manual quality inspection and further improve the accuracy and efficiency of policy quality inspection, in recent years, attempts have been made to directly use machine learning and data mining methods to model the original policy data. However, this method also faces many challenges. First, policy data is high-dimensional and sparse, that is, the data contains a large number of features and attributes, but many of these features and attributes are not of practical value for risk identification, which increases the complexity and computational cost of model training. Secondly, feature engineering is a key link in machine learning and data mining, which determines whether the model can effectively identify risk points. However, feature engineering of policy data is complex and cumbersome, requiring professional knowledge and experience, which is a huge challenge for non-professionals.

[0005] In summary, the traditional policy quality inspection methods have problems such as low efficiency and serious waste of human resources, and the direct application of machine learning methods in existing technologies also has its own limitations. Therefore, how to efficiently and accurately identify high-risk anomalies in policies has become a technical problem that needs to be solved in the insurance business. Summary of the invention

[0006] The purpose of the embodiments of the present application is to propose a policy quality inspection method, device, computer equipment and storage medium to solve the problem of being unable to accurately and effectively evaluate and analyze the risks of insurance policies.

[0007] In order to solve the above technical problems, the embodiment of the present application provides a policy quality inspection method, which adopts the following technical solution:

[0008] Obtaining insurance policy data, and extracting key insurance policy basic information features, insured person features, and insurance type features from the insurance policy data;

[0009] Building a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type;

[0010] Performing policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map;

[0011] Extracting high-risk insurance policies from the risk distribution map, and performing cluster analysis on the high-risk insurance policies to obtain risk analysis information;

[0012] Generate a policy quality inspection report based on the risk analysis information.

[0013] Furthermore, the step of obtaining the insurance policy data and extracting key insurance policy basic information characteristics, insured person characteristics, and insurance type characteristics from the insurance policy data specifically includes:

[0014] Obtaining insurance policy data from a database, and extracting structured data and unstructured data from the insurance policy data;

[0015] Preprocessing the structured data and the unstructured data to obtain standard structured data and standard unstructured data;

[0016] Feature extraction is performed on the standard structured data and the standard unstructured data to obtain the basic information features of the insurance policy, the features of the insured, and the features of the insurance type.

[0017] Furthermore, the step of constructing a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type specifically includes:

[0018] Quantifying the basic information features of the insurance policy, the features of the insured, and the features of the insurance type to obtain quantitative features of the insurance policy, quantitative features of the insured, and quantitative features of the insurance type;

[0019] Constructing a policy complexity evaluation function, a risk exposure evaluation function, and an abnormal index evaluation function based on the policy quantitative characteristics, the insured quantitative characteristics, and the insurance type quantitative characteristics;

[0020] The policy complexity assessment function, risk exposure assessment function and abnormal index assessment function are integrated into the risk assessment model.

[0021] Furthermore, the step of performing policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map specifically includes:

[0022] extracting insurance policy risk assessment indicators from the insurance policy data;

[0023] Mapping the insurance policy risk assessment index to the multidimensional space of the risk assessment model to construct an ant search space;

[0024] Initialize ant colony algorithm parameters, search in the ant search space according to the ant colony algorithm parameters, obtain pheromone concentration results, and use the pheromone concentration results as policy risk assessment results;

[0025] The risk distribution map is generated according to the policy risk assessment result.

[0026] Furthermore, the step of extracting high-risk policies from the risk distribution map and performing cluster analysis on the high-risk policies to obtain risk analysis information specifically includes:

[0027] Identify the risk level of the risk distribution map, and extract the high-risk insurance policy from the insurance policy data according to the identification result;

[0028] Extracting features of the high-risk insurance policy to obtain high-risk insurance policy features;

[0029] Performing cluster analysis on the high-risk insurance policies according to a preset clustering algorithm and the characteristics of the high-risk insurance policies to obtain high-risk insurance policy clusters of different categories;

[0030] Classify the problems of the high-risk policy cluster according to the association rule mining algorithm and generate a risk problem list;

[0031] The risk analysis information is obtained by performing reasoning analysis on the risk problem list according to a pre-established business knowledge base.

[0032] Furthermore, the step of identifying the risk level of the risk distribution map and extracting the high-risk insurance policy from the insurance policy data according to the identification result specifically includes:

[0033] Acquire an ant colony algorithm path data set based on the risk distribution map, and preprocess the ant colony algorithm path data set to obtain a standard path data set;

[0034] Obtaining preset risk policy characteristic indicators, and extracting abnormal premium ratio information, frequent claims number information, and multiple protection coverage information from the standard path data set according to the risk policy characteristic indicators;

[0035] Inputting the abnormal premium ratio information, frequent claims number information, and multiple protection coverage information into the pre-trained risk identification model to obtain a risk level identification result;

[0036] The high-risk insurance policy is extracted from the insurance policy data according to the risk level identification result.

[0037] Furthermore, the step of generating a policy quality inspection report based on the risk analysis information specifically includes:

[0038] Obtaining a template extraction identifier, and extracting a quality inspection report template from a database according to the template extraction identifier;

[0039] Analyze the risk analysis information to obtain risk cause information, risk impact information, and risk solution;

[0040] Optimizing the risk solution based on a decision tree algorithm to obtain an optimized risk solution;

[0041] Fill the risk cause information, the risk impact information, and the optimized risk solution into the quality inspection report template to generate the policy quality inspection report.

[0042] In order to solve the above technical problems, the embodiment of the present application further provides a policy quality inspection device, which adopts the following technical solution:

[0043] A feature extraction module is used to obtain policy data and extract key policy basic information features, insured person features, and insurance type features from the policy data;

[0044] A model building module, used to build a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type;

[0045] A risk assessment module, used to perform policy risk assessment on the policy data based on an ant colony algorithm and the risk assessment model to obtain a risk distribution map;

[0046] A risk analysis module, used to extract high-risk policies from the risk distribution map and perform cluster analysis on the high-risk policies to obtain risk analysis information;

[0047] A report generation module is used to generate a policy quality inspection report based on the risk analysis information.

[0048] In order to solve the above technical problems, the embodiment of the present application further provides a computer device, which adopts the following technical solution:

[0049] A computer device includes a memory and a processor, wherein the memory stores computer-readable instructions, and when the processor executes the computer-readable instructions, the steps of the insurance policy quality inspection method as described in any one of the above items are implemented.

[0050] In order to solve the above technical problems, the embodiment of the present application further provides a computer-readable storage medium, which adopts the following technical solution:

[0051] A computer-readable storage medium having computer-readable instructions stored thereon, wherein the computer-readable instructions, when executed by a processor, implement the steps of the insurance policy quality inspection method as described in any one of the above items.

[0052] Compared with the prior art, the embodiment of the present application has the following beneficial effects: by acquiring the policy data, the embodiment extracts the key basic information characteristics of the policy, the characteristics of the insured, and the characteristics of the insurance type from the policy data; constructs a risk assessment model based on the basic information characteristics of the policy, the characteristics of the insured, and the characteristics of the insurance type; performs policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map; extracts high-risk policies from the risk distribution map, and performs cluster analysis on the high-risk policies to obtain risk analysis information; generates a policy quality inspection report based on the risk analysis information. This effectively realizes accurate and effective risk assessment and analysis of the policy, and generates a corresponding quality inspection report for display. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the scheme in the present application, a brief introduction is given below to the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 is an exemplary system architecture diagram to which the present application may be applied;

[0055] Figure 2 A flow chart of an embodiment of the insurance policy quality inspection method according to the present application;

[0056] Figure 3 yes Figure 2 A flowchart of a specific implementation of step S10;

[0057] Figure 4 yes Figure 2 A flowchart of a specific implementation of step S20;

[0058] Figure 5 yes Figure 2A flowchart of a specific implementation of step S30;

[0059] Figure 6 yes Figure 2 A flowchart of a specific implementation of step S40;

[0060] Figure 7 yes Figure 6 A flowchart of a specific implementation of step S401;

[0061] Figure 8 yes Figure 2 A flowchart of a specific implementation of step S50;

[0062] Fig. 9 It is a structural schematic diagram of an embodiment of a policy quality inspection device according to the present application;

[0063] Fig.10 It is a structural diagram of an embodiment of a computer device according to the present application. DETAILED DESCRIPTION

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of the present application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application; the terms "including" and "having" and any variations thereof in the specification and claims of the present application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of the present application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0065] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor does it refer to non-related or alternative embodiments that are mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0066] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

[0067] like Figure 1As shown, the system architecture 100 may include a terminal device 101, a network 102 and a server 103. The terminal device 101 may be a laptop 1011, a tablet computer 1012 or a mobile phone 1013. The network 102 is used to provide a medium for a communication link between the terminal device 101 and the server 103. The network 102 may include various connection types, such as wired, wireless communication links or optical fiber cables.

[0068] The user can use the terminal device 101 to interact with the server 103 through the network 102 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, such as web browser applications, shopping applications, search applications, instant messaging tools, email clients, social platform software, etc.

[0069] The terminal device 101 can be any electronic device with a display screen and supporting web browsing. In addition to a laptop computer 1011, a tablet computer 1012 or a mobile phone 1013, the terminal device 101 can also be an e-book reader, an MP3 player (Moving Picture Experts Group Audio Layer III), an MP4 (Moving Picture Experts Group Audio Layer IV), a laptop computer, a desktop computer, etc.

[0070] The server 103 may be a server that provides various services, such as a background server that provides support for a web page displayed on the terminal device 101 .

[0071] It should be noted that the insurance policy quality inspection method provided in the embodiment of the present application is generally executed by a server / terminal device, and accordingly, the insurance policy quality inspection device is generally set in the server / terminal device.

[0072] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0073] Continue to refer Figure 2 , shows a flow chart of an embodiment of a method for quality inspection of insurance policies according to the present application. The method for quality inspection of insurance policies comprises the following steps:

[0074] Step S10, obtaining insurance policy data, and extracting key insurance policy basic information features, insured person features, and insurance type features from the insurance policy data;

[0075] In this embodiment, the policy data can be obtained from the database and file server of the insurance company. The policy data includes structured data and unstructured data. Key policy basic information characteristics, insured person characteristics, and insurance type characteristics are extracted from the above structured data and unstructured data. The policy basic information characteristics include: policy number, insurance type, insurance amount, insurance period, premium, insurance date, and payment method. The insured person characteristics include: name, gender, age, occupation, health status (based on the health declaration in the unstructured data), and income level. The insurance type characteristics include: insurance type (life insurance, car insurance, health insurance, etc.), insurance subtype (such as critical illness insurance and medical insurance in health insurance), risk level, coverage (such as hospitalization expenses and outpatient expenses in medical insurance), and specific terms (such as deductibles and compensation ratios).

[0076] Step S20, constructing a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type;

[0077] In this embodiment, the risk assessment model is used to assess the risk level corresponding to the insurance policy by quantifying the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type, and constructing the insurance policy complexity assessment function, risk exposure assessment function, and abnormal indicator assessment function based on the quantified characteristics, thereby generating a risk assessment model that can effectively assess the complexity of the insurance policy, the risk exposure, and the abnormal indicators.

[0078] Step S30, performing policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map;

[0079] In this embodiment, the risk distribution map is a graphical information generated by the path search results obtained by the ant colony algorithm in the multidimensional space of the risk assessment model. The path search results correspond to the assessment results of the risk assessment model, that is, the risk results corresponding to each information in the assessment policy. Among them, the ant colony algorithm (Ant Colony Optimization, ACO) regards each dimension of the risk assessment model as a coordinate axis in a multidimensional space, and regards different risk points as points in this space. The ant colony algorithm simulates the movement and path selection process of ants in the multidimensional space to find the optimal path from the starting point to the end point. The optimal path in the path search results can be regarded as the path with the lowest risk in the risk assessment model. By comparing the pheromone concentration (i.e., the risk size) on different paths, the risk level of each path can be evaluated, and a risk distribution map can be constructed according to the risk level of each path.

[0080] Step S40, extracting high-risk policies from the risk distribution map, and performing cluster analysis on the high-risk policies to obtain risk analysis information;

[0081] In this embodiment, high-risk policies refer to policy data that is determined to be high-risk after evaluation, and can be determined by indicators corresponding to high-risk policies. Risk analysis information is information about the risk causes, impacts and solutions of high-risk policies obtained after clustering and knowledge reasoning based on high-risk policies. By extracting corresponding information features from the path data set corresponding to the above-mentioned ant colony algorithm, feature information for identifying high-risk policies is obtained, and then the feature information is input into the pre-trained risk identification model to obtain the risk level identification result. Among them, the path data set corresponding to the ant colony algorithm refers to the policy data for path search in the risk assessment model using the ant colony algorithm.

[0082] Step S50: Generate a policy quality inspection report based on the risk analysis information.

[0083] In this embodiment, a policy quality inspection report is generated based on the information on the risk causes, impacts and solutions of high-risk policies in the risk analysis information and a preset quality inspection report template. The content of the policy quality inspection report corresponds to risk cause information, risk impact information and optimized risk solution, so as to provide effective information support and reference for subsequent review.

[0084] This embodiment obtains insurance policy data, extracts key insurance policy basic information features, insured person features, and insurance type features from the insurance policy data; builds a risk assessment model based on the insurance policy basic information features, the insured person features, and the insurance type features; performs insurance policy risk assessment on the insurance policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map; extracts high-risk insurance policies from the risk distribution map, and performs cluster analysis on the high-risk insurance policies to obtain risk analysis information; generates an insurance policy quality inspection report based on the risk analysis information. This effectively achieves accurate and effective risk assessment and analysis of insurance policies, and generates corresponding quality inspection reports for display.

[0085] refer to Figure 3 In some optional implementations of this embodiment, step S10 includes the following steps:

[0086] Step S101, obtaining insurance policy data from a database, and extracting structured data and unstructured data from the insurance policy data;

[0087] In this embodiment, the identification can be extracted from the policy data and matched and queried in the database, and the matched policy data can be extracted to obtain. Structured data refers to data with a specific structure, and this part of the data usually exists in the form of a table, such as policy number, insured's name, date of birth, insurance amount, insurance period, etc. Unstructured data refers to data without a specific structure, such as text data like policy description, insured's health declaration, claim report, etc., or multimedia data such as pictures and scanned copies.

[0088] Step S102, preprocess the structured data and the unstructured data to obtain standard structured data and standard unstructured data;

[0089] In this embodiment, the preprocessing of structured data includes removing noise data: deleting invalid or outlier values (such as age significantly exceeding the reasonable range, insurance amount with a negative value, etc.); filling in missing values: using interpolation methods (such as mean, median, mode), prediction models or business rules to fill in the missing values; data conversion: converting date and categorical data into a format suitable for analysis (such as converting date to timestamp, converting categorical data to numerical encoding). The preprocessing of unstructured data includes text data: using natural language processing tools for word segmentation; removing common but meaningless words for analysis (such as "de", "shi"); restoring words to their basic forms (such as restoring "running" to "run"); sentiment analysis: evaluating the sentiment tendency (positive, negative, neutral) of the text. Multimedia data: For images, text information can be extracted through OCR (Optical Character Recognition) technology, or image classification and recognition can be performed.

[0090] Step S103, extract features from the standard structured data and the standard unstructured data to obtain the basic policy information features, the insured features, and the insurance type features.

[0091] In this embodiment, the feature extraction of standard structured data can be achieved by querying the required field information, mapping the query result of the field information to the corresponding feature variables, and performing data conversion, such as converting the date field to the standard format, and normalizing or discretizing the numerical field. The feature extraction of standard unstructured data can be achieved by using natural language processing technology to perform word segmentation, standardization, entity naming, etc. on the text data, and then converting the text data into a numerical vector through text representation methods such as Word2Vec; extracting key features in the image, such as shape, color, texture, etc. through image recognition algorithms.

[0092] This embodiment obtains policy data from a database, and extracts structured data and unstructured data from the policy data; preprocesses the structured data and the unstructured data to obtain standard structured data and standard unstructured data; and performs feature extraction on the standard structured data and the standard unstructured data, thereby effectively obtaining basic policy information features, insured person features, and insurance type features that characterize various information in the policy, so as to facilitate the subsequent construction of a risk assessment model.

[0093] refer to Figure 4 In some optional implementations of this embodiment, step S20 includes the following steps:

[0094] Step S201, quantifying the dimensional features of the basic information features of the insurance policy, the features of the insured, and the features of the insurance type, to obtain quantitative features of the insurance policy, quantitative features of the insured, and quantitative features of the insurance type;

[0095] In this embodiment, the quantification of the basic information features of the insurance policy includes: converting the classification features such as the insurance policy type and the insurance period into numerical features through encoding (such as unique hot encoding and label encoding); standardizing or normalizing the continuous features such as the premium amount to ensure the comparability between different features. The quantification of the insured's characteristics includes: converting the classification features such as gender into numerical features, such as 1 for male and 0 for female; segmenting or standardizing the continuous features such as age; using text mining technology to convert text descriptions such as health status into numerical features, such as the score of the severity of the disease; and constructing risk level assessment indicators based on factors such as the insured's occupation and income level. The quantification of the characteristics of insurance types includes: classifying and encoding the types of insurance types; constructing numerical indicators reflecting the risk characteristics of insurance types based on the coverage and compensation conditions of the insurance types; analyzing historical data and calculating statistical indicators such as the average compensation rate and accident rate of each insurance type as the basis for quantifying the risk of the insurance type.

[0096] Step S202, constructing a policy complexity evaluation function, a risk exposure evaluation function, and an abnormal index evaluation function based on the policy quantitative characteristics, the insured quantitative characteristics, and the insurance type quantitative characteristics;

[0097] In this embodiment, multiple dimensional features such as policy type, insurance period, premium amount, insured characteristics, and insurance type characteristics are integrated into a comprehensive index through weighted summation, principal component analysis (PCA), cluster analysis, etc., to construct a quantitative index of policy complexity, and then generate a policy complexity evaluation function based on the quantitative index. Using machine learning algorithms such as logistic regression and decision trees, the risk exposure is predicted and evaluated based on factors such as the age, health status, and occupation of the insured, as well as the coverage and compensation conditions of the insurance type, to construct a quantitative index of risk exposure, and then generate a risk exposure evaluation function based on the quantitative index. By using statistical methods (such as Z scores, box plots, etc.) or machine learning algorithms (such as isolation forests, one-class support vector machines, etc.) for anomaly detection, abnormal features in policies, insured persons, or insurance types are identified, such as abnormally high or low premiums, abnormal insured age, etc., to construct an abnormal index evaluation function for detecting whether there are similar abnormal situations in new policies.

[0098] Step S203, integrating the policy complexity assessment function, risk exposure assessment function, and abnormal index assessment function into the risk assessment model.

[0099] In this embodiment, the policy complexity assessment function, the risk exposure assessment function, and the abnormal index assessment function are integrated into the same model to generate a risk assessment model. In the risk assessment model, each function is set and has corresponding different weights.

[0100] This embodiment quantifies the dimensional features of the basic information features of the insurance policy, the insured features, and the insurance type features to obtain the quantitative features of the insurance policy, the quantitative features of the insured, and the quantitative features of the insurance type; constructs the policy complexity assessment function, the risk exposure assessment function, and the abnormal index assessment function based on the quantitative features of the insurance policy, the quantitative features of the insured, and the quantitative features of the insurance type; and integrates the policy complexity assessment function, the risk exposure assessment function, and the abnormal index assessment function into the risk assessment model. This effectively implements the construction of a risk assessment model that effectively assesses the risk level of the insurance policy data based on the above-mentioned basic information features of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type, so as to ensure the accuracy and effectiveness of the risk assessment and facilitate the subsequent risk assessment of the insurance policy.

[0101] refer to Figure 5 In some optional implementations of this embodiment, step S30 includes the following steps:

[0102] Step S301, extracting insurance policy risk assessment indicators from the insurance policy data;

[0103] In this embodiment, the policy data may include basic policy information (such as policy number, information of the policyholder, information of the insured, insurance type, insurance period, premium amount, etc.), historical claims records, the insured's health status, occupation, income level, etc. The policy risk assessment indicators include the insured's age, gender, health status score, occupational risk level, risk level of the insurance type, historical claims ratio, etc. The policy risk assessment indicators may be extracted from the policy data by using preset risk assessment indicator extraction rules.

[0104] Step S302, mapping the insurance policy risk assessment index to the multidimensional space of the risk assessment model to construct an ant search space;

[0105] In this embodiment, the extracted risk assessment indicators are mapped to the multidimensional space of the risk assessment model, each dimension in the multidimensional space represents a risk assessment indicator, and a suitable search space is defined according to the scope and number of the risk assessment indicators. The space is used as the search domain of the ant colony algorithm to effectively construct the ant search space.

[0106] Step S303, initializing ant colony algorithm parameters, searching in the ant search space according to the ant colony algorithm parameters, obtaining pheromone concentration results, and using the pheromone concentration results as policy risk assessment results;

[0107] In this embodiment, initializing the parameters of the ant colony algorithm includes setting the number of ants in the ant colony, which needs to be adjusted according to the scale and complexity of the actual insurance policy data; setting the initial concentration of pheromones, which determines the initial preference of the algorithm search; setting the volatility coefficient of pheromones, which determines the speed at which pheromones decay over time and affects the exploration and utilization capabilities of the algorithm; setting parameters for calculating the probability of ant transfer, such as heuristic factors, pheromone importance factors, etc. By randomly generating the initial positions of ants in the ant search space, where each position represents a set of risk assessment index values, and then calculating the transfer probability of the ant to the next position based on the risk assessment index value, pheromone concentration and transfer probability calculation parameters at the current position, the next search position is selected based on the transfer probability, and the position information of the ant is updated. The search continues until the ant reaches the risk point position, and the pheromone concentration result corresponding to the path is obtained.

[0108] Step S304, generating the risk distribution map according to the insurance policy risk assessment result.

[0109] In this embodiment, the policy risk assessment results correspond to the pheromone concentration results of the above-mentioned multiple paths. The policy risk assessment results reflect the risk situation of each policy. The policy risk assessment results are combined with visualization tools to generate a risk distribution map that effectively reflects the risk distribution of the policy data.

[0110] This embodiment extracts insurance policy risk assessment indicators from the insurance policy data; maps the insurance policy risk assessment indicators to the multidimensional space of the risk assessment model to construct an ant search space; initializes ant colony algorithm parameters, searches in the ant search space according to the ant colony algorithm parameters, obtains pheromone concentration results, and uses the pheromone concentration results as insurance policy risk assessment results; generates the risk distribution map according to the insurance policy risk assessment results. This effectively realizes the generation of a risk distribution map that effectively reflects the specific distribution of information risks based on the insurance policy data, so as to facilitate further analysis based on the risk distribution map.

[0111] Continue to refer Figure 6 In some optional implementations of this embodiment, step S40 includes the following steps:

[0112] Step S401, identifying the risk level of the risk distribution map, and extracting the high-risk insurance policies from the insurance policy data according to the identification result;

[0113] In this embodiment, a high-risk policy refers to policy data in which the abnormal premium ratio information, frequent claims number information, and multiple protection coverage information are higher than the preset judgment standards, wherein the abnormal premium ratio information, frequent claims number information, and multiple protection coverage information are judged based on a pre-trained risk identification model, and the abnormal premium ratio information, frequent claims number information, and multiple protection coverage information correspond to different weights respectively, and the weighted total score is used to comprehensively judge whether the policy data is a high-risk policy.

[0114] Step S402, extracting features of the high-risk insurance policy to obtain high-risk insurance policy features.

[0115] In this embodiment, high-risk policy features may include the insured's age, gender, health status, occupation, insurance type, insurance period, premium amount, etc. The extracted high-risk policy features may be processed by missing value filling, outlier processing, feature scaling, etc. to ensure the quality and consistency of the feature data and facilitate subsequent feature application.

[0116] Step S403, performing cluster analysis on the high-risk insurance policies according to a preset clustering algorithm and the characteristics of the high-risk insurance policies to obtain high-risk insurance policy clusters of different categories;

[0117] In this embodiment, the preset clustering algorithm adopts the K-means algorithm. The application of the K-means algorithm includes the setting of algorithm parameters: according to the characteristics of high-risk policies and business requirements, the number of clusters K is preset. The selection of the K value can be determined by trial and error, elbow method or silhouette coefficient method; K initial centroids are randomly selected, and these centroids will be used as the starting point of clustering. Iterative clustering process: calculate the distance between each high-risk policy feature vector and the K centroids, and assign each policy to the cluster with the nearest centroid; recalculate the centroid of each cluster, which is usually the mean of all policy feature vectors in the cluster; repeat the above two steps until the position of the centroid no longer changes significantly or reaches the preset number of iterations, and output the policy clustering results at this time as high-risk policy clusters of different categories.

[0118] Step S404, classifying the problems of the high-risk policy cluster according to the association rule mining algorithm, and generating a risk problem list;

[0119] In this embodiment, the association rule mining algorithm adopts the Apriori algorithm, and the Apriori algorithm is used to discover frequent item sets and association rules in the data. Appropriate support and confidence thresholds can be set to screen out meaningful association rules. By analyzing these association rules, the characteristics or factors related to high-risk policies are identified to obtain association rule mining results. According to the results of association rule mining, the problems in the high-risk policy cluster are classified, such as policyholder fraud, poor health of the insured, excessive insurance amount, and excessive insurance period, so as to generate a risk problem list, listing the specific problems and related policies under each category.

[0120] Step S405: perform reasoning analysis on the risk problem list according to the pre-established business knowledge base to obtain the risk analysis information.

[0121] In this embodiment, the pre-established business knowledge base is a pre-established database containing insurance business knowledge, which includes information such as insurance clauses, compensation policies, industry regulations, historical cases, etc. For each problem in the risk problem list, the business knowledge base is combined with the knowledge reasoning algorithm to perform reasoning analysis to identify the root cause of the problem, potential risk points, and possible solutions or preventive measures to obtain risk analysis information.

[0122] This embodiment identifies the risk level of the risk distribution map, and extracts the high-risk policies from the policy data based on the identification results; extracts features of the high-risk policies to obtain high-risk policy features; clusters the high-risk policies according to a preset clustering algorithm and the high-risk policy features to obtain high-risk policy clusters of different categories; classifies the problems of the high-risk policy clusters according to an association rule mining algorithm to generate a risk problem list; and infers and analyzes the risk problem list based on a pre-established business knowledge base, thereby effectively obtaining risk analysis information for association analysis of the high-risk policy set business knowledge, so as to facilitate the subsequent generation of a policy quality inspection report based on the risk analysis information.

[0123] Continue to refer Figure 7 In some optional implementations of this embodiment, step S401 includes the following steps:

[0124] Step S4011, acquiring an ant colony algorithm path data set based on the risk distribution map, and preprocessing the ant colony algorithm path data set to obtain a standard path data set;

[0125] In this embodiment, the ant colony algorithm path data set refers to the insurance policy data for path search in the risk assessment model using the ant colony algorithm. Preprocessing includes data cleaning (removing invalid paths, correcting erroneous data) and data standardization (risk value standardization, path length standardization). By performing the above preprocessing on the ant colony algorithm path data set, a standard path data set is effectively obtained.

[0126] Step S4012, obtaining a preset risk policy characteristic index, and extracting abnormal premium ratio information, frequent claims information, and multiple protection coverage information from the standard path data set according to the risk policy characteristic index;

[0127] In this embodiment, the preset risk policy characteristic indicators include abnormal premium ratio, frequent claims, and multiple protection coverage, wherein the abnormal premium ratio refers to the ratio of the premium amount exceeding the normal range, the frequent claims refers to the number of claims occurring within a certain period of time exceeding the preset threshold, and the multiple protection coverage refers to whether the types and scopes of protection provided in the policy are excessively overlapping or insufficient. The abnormal premium ratio information is extracted by searching the fields related to the premium in the standard path data set, and calculating the premium ratio of each policy, that is, the ratio of the premium of the policy to the average premium or the preset premium threshold, and comparing the calculated premium ratio with the preset ratio threshold to determine whether the premium ratio is abnormal. The frequent claims information is extracted by searching the fields related to the claims in the standard path data set, and counting the number of claims within a certain time window, and judging whether the number of claims exceeds the preset number threshold to determine whether the number of claims is frequent. The extraction of multiple protection coverage information is carried out by searching for fields related to the types and scope of protection in the standard path data set, analyzing the types and scope of protection provided, identifying whether there is excessive overlap or insufficient coverage, and recording multiple protection coverage information based on the analysis results, including the specific content of excessive coverage and insufficient coverage.

[0128] Step S4013, inputting the abnormal premium ratio information, frequent claims number information, and multiple protection coverage information into a pre-trained risk identification model to obtain a risk level identification result;

[0129] In this embodiment, the pre-trained risk identification model can adopt a random forest model, and the random forest model is trained and optimized through historical sample data so that the random forest model can be used for risk assessment tasks of high-risk insurance policies. The risk level identification result includes risk level and risk probability. In this embodiment, the risk level includes primary risk, secondary risk, and tertiary risk. The risk probability corresponds to the probability value of each insurance policy belonging to a certain risk level to provide explanatory risk assessment information.

[0130] Step S4014: extracting the high-risk insurance policy from the insurance policy data according to the risk level identification result.

[0131] In this embodiment, level one risk is a low-level risk, level two risk is a medium-level risk, and level three risk is a high-level risk. Insurance policies with a risk level identification result of level three risk are identified and extracted from the insurance policy data to effectively obtain high-risk insurance policies.

[0132] This embodiment obtains an ant colony algorithm path data set based on the risk distribution map, and pre-processes the ant colony algorithm path data set to obtain a standard path data set; obtains preset risk policy characteristic indicators, and extracts abnormal premium ratio information, frequent claims information, and multiple protection coverage information from the standard path data set according to the risk policy characteristic indicators; inputs the abnormal premium ratio information, frequent claims information, and multiple protection coverage information into a pre-trained risk identification model to obtain a risk level identification result; extracts the high-risk policy from the policy data according to the risk level identification result. This enables accurate extraction of high-risk policies that meet the preset risk policy characteristic indicator requirements from the policy data, so as to facilitate subsequent feature extraction processing.

[0133] Continue to refer Figure 8 In some optional implementations of this embodiment, step S50 includes the following steps:

[0134] Step S501, obtaining a template extraction identifier, and extracting a quality inspection report template from a database according to the template extraction identifier;

[0135] In this embodiment, the template extraction identifier is a unique identifier corresponding to the quality inspection report template. A matching query is performed in the database according to the template extraction identifier to query the quality inspection report template and extract it.

[0136] Step S502, parsing the risk analysis information to obtain risk cause information, risk impact information, and risk solution;

[0137] In this embodiment, the risk analysis information includes risk cause information, risk impact information, and risk solution, wherein the risk cause information refers to the various causes and factors that lead to the occurrence of risk events. For high-risk policies, the risk causes involve multiple aspects, such as the personal characteristics of the policyholder or the insured (such as age, health status, occupation, etc.), the design of the policy terms (such as the insurance amount, insurance period, compensation conditions, etc.), and changes in the external environment (such as economic cycles, adjustments to laws and regulations, natural disasters, etc.). Risk impact information refers to the possible consequences and impacts after the occurrence of a risk event. For high-risk policies, the risk impacts include economic losses (such as an increase in compensation amount, a decrease in premium income, etc.), a decrease in customer satisfaction (such as an increase in complaints due to improper compensation handling), and a damaged brand image (such as negative public opinion caused by the exposure of the risk event by the media). Risk solutions refer to measures and methods developed to reduce the probability of risk occurrence and mitigate the impact of risk based on the identified risk causes and impacts. For high-risk policies, risk solutions include adjusting policy terms (such as raising the compensation threshold, shortening the insurance period, etc.), strengthening risk control (such as raising underwriting standards, strengthening risk monitoring, etc.), and optimizing business processes (such as improving claims processing efficiency, strengthening customer service, etc.).

[0138] Step S503, optimizing the risk solution based on a decision tree algorithm to obtain an optimized risk solution;

[0139] In this embodiment, the decision tree algorithm can use CART to optimize the risk solution. By viewing the branches and nodes of the decision tree, the key factors and paths leading to high-risk policies can be identified. These paths reflect the probability and impact of risk under different feature combinations. Based on the analysis results of the decision tree, targeted risk solution optimization strategies can be formulated. For example, for certain specific solutions, it may be necessary to adjust the policy terms, strengthen risk control measures, or optimize business processes.

[0140] Step S504, fill the risk cause information, the risk impact information, and the optimized risk solution into the quality inspection report template to generate the policy quality inspection report.

[0141] In this embodiment, the quality inspection report template has preset filling positions, filling methods and layout methods corresponding to the risk cause information, risk impact information and optimized risk solution. The risk cause information, risk impact information and optimized risk solution are filled into the quality inspection report template according to the filling positions, filling methods and layout methods to generate a policy quality inspection report for subsequent analysis and processing.

[0142] In this embodiment, a corresponding policy quality inspection report can also be generated for the non-high-risk policies in the risk distribution map. The policy quality inspection report corresponding to the non-high-risk policies may include the policy risk situation. The policy risk situation may be the risk distribution situation of the policy. By generating a policy quality inspection report corresponding to the non-high-risk policies, the risk situation of the non-high-risk policies can be visualized for subsequent analysis.

[0143] This embodiment obtains a template extraction identifier, extracts a quality inspection report template from a database according to the template extraction identifier; parses the risk analysis information to obtain risk cause information, risk impact information, and risk solution; optimizes the risk solution based on a decision tree algorithm to obtain an optimized risk solution; fills the risk cause information, risk impact information, and optimized risk solution into the quality inspection report template to generate the policy quality inspection report. This effectively realizes the generation of a policy quality inspection report based on the comprehensive risk cause information, risk impact information, and risk solution, so as to improve the reliability of the policy quality inspection report.

[0144] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0145] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0146] Further references Fig. 9 , as a response to the above Figure 1 In order to realize the method shown in the figure, the present application provides an embodiment of a policy quality inspection device, which is similar to Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0147] like Fig. 9 As shown, the insurance policy quality inspection device 800 described in this embodiment includes: a feature extraction module 601, a model construction module 602, a risk assessment module 603, a risk analysis module 604, and a report generation module 605. Among them:

[0148] The feature extraction module 601 is used to obtain the policy data and extract the key policy basic information features, the insured person features, and the insurance type features from the policy data;

[0149] A model building module 602 is used to build a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type;

[0150] The risk assessment module 603 is used to perform policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map;

[0151] The risk analysis module 604 is used to extract high-risk policies from the risk distribution map and perform cluster analysis on the high-risk policies to obtain risk analysis information;

[0152] The report generation module 605 is used to generate a policy quality inspection report based on the risk analysis information.

[0153] By adopting the above-mentioned insurance policy quality inspection device, this embodiment can obtain insurance policy data, extract key insurance policy basic information characteristics, insured person characteristics, and insurance type characteristics from the insurance policy data; construct a risk assessment model based on the insurance policy basic information characteristics, the insured person characteristics, and the insurance type characteristics; perform insurance policy risk assessment on the insurance policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map; extract high-risk insurance policies from the risk distribution map, and perform cluster analysis on the high-risk insurance policies to obtain risk analysis information; generate an insurance policy quality inspection report based on the risk analysis information. This effectively realizes accurate and effective risk assessment and analysis of insurance policies, and generates corresponding quality inspection reports for display.

[0154] To solve the above technical problems, the present application also provides a computer device. Fig.10 , Fig.10 This is a basic structural block diagram of the computer device in this embodiment.

[0155] The computer device 7 includes a memory 71, a processor 72, and a network interface 73 that are interconnected through a system bus. It should be noted that the figure only shows a computer device 7 with components 91-73, but it should be understood that it is not required to implement all the components shown, and more or fewer components can be implemented instead. Among them, those skilled in the art can understand that the computer device here is a device that can automatically perform numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to microprocessors, application specific integrated circuits (Application Specific Integrated Circuit, ASIC), programmable gate arrays (Field-Programmable Gate Array, FPGA), digital processors (Digital Signal Processor, DSP), embedded devices, etc.

[0156] The computer device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The computer device may interact with a user through a keyboard, a mouse, a remote controller, a touch pad, or a voice control device.

[0157] The memory 71 includes at least one type of readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 71 can be an internal storage unit of the computer device 7, such as a hard disk or memory of the computer device 7. In other embodiments, the memory 71 can also be an external storage device of the computer device 7, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (FlashCard), etc. equipped on the computer device 7. Of course, the memory 71 can also include both the internal storage unit of the computer device 7 and its external storage device. In this embodiment, the memory 71 is generally used to store the operating system and various application software installed on the computer device 7, such as computer-readable instructions of the policy quality inspection method, etc. In addition, the memory 71 can also be used to temporarily store various types of data that have been output or are to be output.

[0158] The processor 72 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 72 is generally used to control the overall operation of the computer device 7. In this embodiment, the processor 72 is used to run computer-readable instructions stored in the memory 71 or process data, such as computer-readable instructions for running the policy quality inspection method.

[0159] The network interface 73 may include a wireless network interface or a wired network interface. The network interface 73 is generally used to establish a communication connection between the computer device 7 and other electronic devices.

[0160] By using the above-mentioned computer equipment, this embodiment can obtain insurance policy data, extract key insurance policy basic information features, insured person features, and insurance type features from the insurance policy data; construct a risk assessment model based on the insurance policy basic information features, the insured person features, and the insurance type features; perform insurance policy risk assessment on the insurance policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map; extract high-risk insurance policies from the risk distribution map, and perform cluster analysis on the high-risk insurance policies to obtain risk analysis information; generate an insurance policy quality inspection report based on the risk analysis information. This effectively realizes accurate and effective risk assessment and analysis of insurance policies, and generates corresponding quality inspection reports for display.

[0161] The present application also provides another embodiment, namely, providing a computer-readable storage medium, which stores computer-readable instructions, and the computer-readable instructions can be executed by at least one processor to enable the at least one processor to perform the steps of the above-mentioned policy quality inspection method.

[0162] By adopting the above-mentioned computer-readable storage medium, this embodiment can obtain insurance policy data, extract key insurance policy basic information characteristics, insured person characteristics, and insurance type characteristics from the insurance policy data; construct a risk assessment model based on the insurance policy basic information characteristics, the insured person characteristics, and the insurance type characteristics; perform insurance policy risk assessment on the insurance policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map; extract high-risk insurance policies from the risk distribution map, and perform cluster analysis on the high-risk insurance policies to obtain risk analysis information; generate an insurance policy quality inspection report based on the risk analysis information. This effectively realizes accurate and effective risk assessment and analysis of insurance policies, and generates corresponding quality inspection reports for display.

[0163] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, 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, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0164] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to perform equivalent replacement of some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. A method for quality inspection of insurance policies, characterized in that: The steps include: Obtaining insurance policy data, and extracting key insurance policy basic information features, insured person features, and insurance type features from the insurance policy data; Building a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type; Performing policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map; Extracting high-risk insurance policies from the risk distribution map, and performing cluster analysis on the high-risk insurance policies to obtain risk analysis information; Generate a policy quality inspection report based on the risk analysis information.

2. The policy quality inspection method according to claim 1, characterized in that: The step of obtaining the policy data and extracting key policy basic information features, insured person features, and insurance type features from the policy data specifically includes: Obtaining insurance policy data from a database, and extracting structured data and unstructured data from the insurance policy data; Preprocessing the structured data and the unstructured data to obtain standard structured data and standard unstructured data; Feature extraction is performed on the standard structured data and the standard unstructured data to obtain the basic information features of the insurance policy, the features of the insured, and the features of the insurance type.

3. The policy quality inspection method according to claim 1, characterized in that: The step of constructing a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type specifically includes: Quantifying the basic information features of the insurance policy, the features of the insured, and the features of the insurance type to obtain quantitative features of the insurance policy, quantitative features of the insured, and quantitative features of the insurance type; Constructing a policy complexity evaluation function, a risk exposure evaluation function, and an abnormal index evaluation function based on the policy quantitative characteristics, the insured quantitative characteristics, and the insurance type quantitative characteristics; The policy complexity assessment function, risk exposure assessment function and abnormal index assessment function are integrated into the risk assessment model.

4. The policy quality inspection method according to claim 1, characterized in that: The step of performing policy risk assessment on the policy data based on the ant colony algorithm and the risk assessment model to obtain a risk distribution map specifically includes: extracting insurance policy risk assessment indicators from the insurance policy data; Mapping the insurance policy risk assessment index to the multidimensional space of the risk assessment model to construct an ant search space; Initialize ant colony algorithm parameters, search in the ant search space according to the ant colony algorithm parameters, obtain pheromone concentration results, and use the pheromone concentration results as policy risk assessment results; The risk distribution map is generated according to the policy risk assessment result.

5. The policy quality inspection method according to claim 1, characterized in that: The step of extracting high-risk policies from the risk distribution map and performing cluster analysis on the high-risk policies to obtain risk analysis information specifically includes: Identify the risk level of the risk distribution map, and extract the high-risk insurance policy from the insurance policy data according to the identification result; Extracting features of the high-risk insurance policy to obtain high-risk insurance policy features; Performing cluster analysis on the high-risk insurance policies according to a preset clustering algorithm and the characteristics of the high-risk insurance policies to obtain high-risk insurance policy clusters of different categories; Classify the problems of the high-risk policy cluster according to the association rule mining algorithm and generate a risk problem list; The risk analysis information is obtained by performing reasoning analysis on the risk problem list according to a pre-established business knowledge base.

6. The policy quality inspection method according to claim 5, characterized in that: The step of identifying the risk level of the risk distribution map and extracting the high-risk insurance policy from the insurance policy data according to the identification result specifically includes: Acquire an ant colony algorithm path data set based on the risk distribution map, and preprocess the ant colony algorithm path data set to obtain a standard path data set; Obtaining preset risk policy characteristic indicators, and extracting abnormal premium ratio information, frequent claims number information, and multiple protection coverage information from the standard path data set according to the risk policy characteristic indicators; Inputting the abnormal premium ratio information, frequent claims number information, and multiple protection coverage information into the pre-trained risk identification model to obtain a risk level identification result; The high-risk insurance policy is extracted from the insurance policy data according to the risk level identification result.

7. The policy quality inspection method according to claim 1, characterized in that: The step of generating a policy quality inspection report based on the risk analysis information specifically includes: Obtaining a template extraction identifier, and extracting a quality inspection report template from a database according to the template extraction identifier; Analyze the risk analysis information to obtain risk cause information, risk impact information, and risk solution; Optimizing the risk solution based on a decision tree algorithm to obtain an optimized risk solution; Fill the risk cause information, the risk impact information, and the optimized risk solution into the quality inspection report template to generate the policy quality inspection report.

8. A policy quality inspection device, characterized in that: include: A feature extraction module is used to obtain policy data and extract key policy basic information features, insured person features, and insurance type features from the policy data; A model building module, used to build a risk assessment model based on the basic information characteristics of the insurance policy, the characteristics of the insured, and the characteristics of the insurance type; A risk assessment module, used to perform policy risk assessment on the policy data based on an ant colony algorithm and the risk assessment model to obtain a risk distribution map; A risk analysis module, used to extract high-risk policies from the risk distribution map and perform cluster analysis on the high-risk policies to obtain risk analysis information; A report generation module is used to generate a policy quality inspection report based on the risk analysis information.

9. A computer device, characterized in that: It includes a memory and a processor, the memory stores computer-readable instructions, and the processor implements the steps of the insurance policy quality inspection method as described in any one of claims 1 to 7 when executing the computer-readable instructions.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the insurance policy quality inspection method as described in any one of claims 1 to 7.

Images