CN118037460A - An insurance product evaluation method and system - Google Patents

Abstract

The invention relates to the technical field of data mining, in particular to an evaluation method and system of an insurance product. The method comprises the following steps: acquiring historical data of insurance products, carrying out claim settlement risk evaluation, and acquiring historical claim settlement risk data; constructing a claim risk evaluation model based on the historical claim risk data; acquiring insurance product data to be evaluated, performing claim risk evaluation, and acquiring claim risk evaluation data; performing coverage evaluation on the historical data of the insurance product to obtain coverage evaluation data; calculating product default interaction frequency of the insurance product data to be evaluated, obtaining product default interaction frequency data to be evaluated, and performing customer interaction experience evaluation to obtain interaction experience evaluation data; and carrying out comprehensive evaluation on the insurance product according to the claim risk evaluation data, the coverage area evaluation data and the interactive experience evaluation data, thereby obtaining an insurance product evaluation report. The invention can improve the accuracy of the insurance product evaluation result.

Description

An insurance product evaluation method and system

Technical Field

The present invention relates to the field of data mining technology, and in particular to an insurance product evaluation method and system.

Background technique

Traditional insurance product evaluation methods are usually based on historical data and statistical models, which are difficult to fully understand customers' personalized needs and risk characteristics. Since these models are usually based on past data and statistical analysis, they may not be able to accurately predict future risk situations, especially for new risks or extreme events. In addition, existing insurance product evaluation systems often only focus on some traditional risk factors, such as gender, occupation, etc., while ignoring many other factors that may affect customer experience, such as customer communication frequency, product payment period, etc. This leads to a lack of comprehensiveness and accuracy in the evaluation results, which cannot fully reflect the personalized risks of customers.

Summary of the invention

Based on this, it is necessary for the present invention to provide an insurance product evaluation method and system to solve at least one of the above technical problems.

To achieve the above purpose, a method for evaluating an insurance product includes the following steps:

Step S1: Obtain historical data of insurance products, and conduct claims risk assessment on the historical data of insurance products, thereby obtaining historical claims risk data; and construct a claims risk assessment model based on the historical claims risk data;

Step S2: Obtain the insurance product data to be evaluated, and perform a claims risk evaluation on the insurance product data to be evaluated using a claims risk evaluation model, thereby obtaining claims risk evaluation data;

Step S3: extracting customer features from historical data of insurance products to obtain historical customer data, and classifying historical customer data by customer loyalty to obtain customer loyalty data; conducting coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data to obtain coverage evaluation data;

Step S4: Calculate the product default interaction frequency for the insurance product data to be evaluated, thereby obtaining the default interaction frequency data for the product to be tested, and perform customer interaction experience evaluation on the default interaction frequency data for the product to be tested and the insurance product historical data, thereby obtaining interaction experience evaluation data;

Step S5: Perform a comprehensive evaluation of the insurance product based on the claims risk evaluation data, coverage evaluation data, and interactive experience evaluation data to obtain an insurance product evaluation report.

The present invention constructs a claims risk assessment model by analyzing the historical claims data of insurance products. Such a model can identify and quantify different types of risks, providing an important basis for subsequent insurance product evaluation. By understanding the historical claims data, the potential claims risks in the future can be better predicted. By performing claims risk assessment on the insurance product data to be evaluated, the risk level of the product can be quantified. This helps insurance companies better understand the risk characteristics of each product, so as to formulate pricing strategies and claims policies more effectively. By extracting customer characteristics and classifying loyalty from historical customer data, customers' behaviors and preferences can be better understood. Such an assessment can help insurance companies better understand their existing customer groups, design more attractive insurance products and services, and improve customer loyalty. By calculating the default interaction frequency of the product and the customer interaction experience assessment, the interactivity and customer friendliness of the insurance product can be evaluated. This helps insurance companies design more attractive and easy-to-use products, improve customer experience, and thus enhance customer satisfaction and loyalty. By comprehensively considering the claims risk assessment, coverage assessment, and interaction experience assessment data, a comprehensive evaluation report of the insurance product can be obtained. Such a comprehensive assessment can help insurance companies fully understand the advantages and room for improvement of the product, and provide an important reference for business decision-making.

Optionally, step S1 specifically includes:

Step S11: Obtaining historical data of insurance products;

Step S12: extracting the term feature of the historical data of insurance products, thereby obtaining historical long-term insurance product data and historical short-term insurance product data;

Step S13: conducting a long-term product claim risk assessment on historical long-term insurance product data, thereby obtaining long-term product claim risk data;

Step S14: conducting a short-term product claim risk assessment on historical short-term insurance product data, thereby obtaining short-term product claim risk data;

Step S15: merging the long-term product claim risk data and the short-term product claim risk data to obtain historical claim risk data;

Step S16: Construct a claims risk assessment model based on historical claims risk data.

By acquiring historical data of insurance products, the present invention can establish a comprehensive data set, which includes past insurance product information, policy details, claims records, etc. This provides basic data for subsequent steps, so that the model can take into account the impact of historical information in the claims risk assessment. Through term feature extraction, the insurance product data can be divided into long-term and short-term, and then the risks of these two types of products can be analyzed in a targeted manner. This helps to more accurately assess the differences in claims risks between long-term and short-term insurance products. By performing claims risk assessment on historical long-term insurance product data, the potential risks of long-term products can be identified and quantified. This helps insurance companies better understand and manage the risks of long-term products and adopt corresponding insurance strategies. Similar to long-term products, performing claims risk assessment on historical short-term insurance product data helps to identify and quantify the risks of short-term products. This enables insurance companies to better understand and respond to the claims risks that short-term products may face. By merging the claims risk data of long-term and short-term products, a more comprehensive historical claims risk data set can be created. This helps to comprehensively consider the risk factors of long-term and short-term products and provide richer information for building a comprehensive claims risk assessment model. A claims risk assessment model is constructed, which is based on historical claims risk data and can predict the claims risks of future insurance products. This provides insurers with a powerful tool to help them price products more accurately, develop risk management strategies, and improve overall business efficiency.

Optionally, step S13 is specifically:

Step S131: extracting features from historical long-term insurance product data, thereby obtaining historical long-term claims data and historical long-term premium payment data;

Step S132: Calculate the claim probability of the same product based on the historical long-term claim data to obtain product claim probability data, and perform statistical analysis on the product claim probability data to obtain the product claim risk threshold;

Step S133: classifying the historical long-term claims data according to the product claims risk threshold, thereby obtaining high claims risk long-term product data and low claims risk long-term product data;

Step S134: Counting the number of missed payments by customers based on historical long-term premium payment data, thereby obtaining missed payment data of high-frequency customers and missed payment data of low-frequency customers;

Step S135: classifying the historical long-term insurance product data by customer risk according to the high-frequency customer missed payment data and the low-frequency customer missed payment data, thereby obtaining high-customer risk long-term product data and low-customer risk long-term product data;

Step S136: performing an intersection operation on the long-term product data with high claim risk and the long-term product data with high customer risk, thereby obtaining the long-term product claim high risk data; performing an intersection operation on the long-term product data with low claim risk and the long-term product data with low customer risk, thereby obtaining the long-term product claim low risk data;

Step S137: Merge the long-term product claim low-risk data and the long-term product claim high-risk data to obtain the long-term product claim risk data.

The present invention can extract useful features from historical data through feature extraction for subsequent risk assessment and classification. The calculation of claim probability and risk threshold can help insurance companies better understand the risk level of products and provide a basis for risk classification. Classifying products according to risk level can help insurance companies formulate risk management strategies more targeted and improve the effectiveness of risk control. Statistics on the number of missed payments by customers can help insurance companies understand customers' payment situations, identify high-risk customers, and take corresponding measures to reduce risks. Classifying customers according to risk levels can help insurance companies better understand the risk characteristics of customer groups and carry out customer management and risk control in a targeted manner. By performing intersection operations on claim risk and customer risk, high-risk products and customers can be more accurately identified, providing more sophisticated data support for risk management. Merging low-risk and high-risk data can obtain a complete risk data set, providing insurance companies with a more comprehensive basis for risk assessment and supporting decisions such as insurance product design and pricing.

Optionally, step S14 is specifically:

Step S141: extracting features from historical short-term insurance product data to obtain historical short-term claims data and customer age data;

Step S142: performing claim frequency statistics on historical short-term claim data, thereby obtaining high-frequency short-term claim data and low-frequency short-term claim data;

Step S143: classify and calculate the customer age data, so as to obtain the age data of older customers and the age data of younger customers;

Step S144: extracting age-restricted insurance products from historical short-term insurance product data based on the age data of senior customers and the age data of junior customers, thereby obtaining short-term age-restricted insurance product data; performing intersection operations on low-frequency short-term claims data and short-term age-restricted insurance product data, thereby obtaining short-term product claims low-risk data;

Step S145: performing customer association on the high-frequency short-term claims data and the age data of senior customers, thereby obtaining first customer association data, and extracting high-risk senior customer product data from the historical short-term insurance product data based on the first customer association data, thereby obtaining high-risk data on senior short-term product claims;

Step S146: performing customer association on the high-frequency short-term claims data and the age data of young customers, thereby obtaining second customer association data, and extracting high-risk young customer product data from the historical short-term insurance product data based on the second customer association data, thereby obtaining high-risk data on claims of young short-term products;

Step S147: Merge the low-risk data on short-term product claims, the high-risk data on senior short-term product claims, and the high-risk data on junior short-term product claims to obtain short-term product claim risk data.

Through feature extraction, the present invention can understand the customer's claims situation and age distribution, and provide a data basis for subsequent risk analysis and customer association. Claim frequency statistics can help identify products with frequent claims, and these data can be regarded as potential high-risk factors. Through the classification of age data, risk analysis and customized insurance product design can be performed for customers of different age groups. By limiting the age range and claim frequency, safer insurance products and customers can be extracted to reduce the risk of claims. Through association analysis, the common characteristics of high-risk customer groups can be discovered, and risk management and product design can be carried out in a targeted manner. By merging data at different risk levels, comprehensive claims risk data can be obtained, providing support for insurance companies to formulate more effective risk management strategies.

Optionally, step S3 specifically includes:

Step S31: extracting customer features from historical insurance product data to obtain historical customer data;

Step S32: extracting the insured product data from the historical customer data, thereby obtaining the customer's insured product data;

Step S33: performing insurance period statistics on the customer insurance product data, thereby obtaining long-term customer insurance product data and short-term customer insurance product data, and calculating the insurance product ratio for the long-term customer insurance product data and short-term customer insurance product data, thereby obtaining customer insurance product ratio data;

Step S34: Count the number of insured products based on the customer insurance data, thereby obtaining the number of insured products for the customer;

Step S35: dividing the customer's insured product ratio data and the number of customer's insured products into customer loyalty categories, thereby obtaining customer loyalty data;

Step S36: Conduct coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data, thereby obtaining coverage evaluation data.

By extracting customer characteristics, the present invention can understand the personal and insurance-related information of historical customers and provide a data basis for subsequent analysis. By extracting customer insurance product data, the customer's insurance purchasing behavior can be understood to form a customer-product association. Understanding the insurance behavior of long-term and short-term customers and calculating the proportion of insured products helps to understand the customer's insurance purchasing preferences and provide guidance for product design and sales strategies. By counting the number of insured products, the customer's insurance portfolio can be understood to provide a reference for product recommendations and pricing. Dividing customers into different loyalty levels can help companies identify high-loyalty customers and low-loyalty customers, thereby formulating corresponding customer relationship management strategies. By evaluating the coverage of the insurance products to be evaluated, the matching degree between the products and customers with different loyalty can be understood, providing data support for product improvement and positioning.

Optionally, the customer loyalty classification in step S35 is specifically as follows:

Conduct product ratio statistical analysis on the customer insurance product ratio data, so as to obtain customer data with a high proportion of short-term products, customer data with a high proportion of long-term products, customer data with monotonous products, and customer data with the same proportion;

Conduct product quantity statistics analysis on the number of insured products purchased by customers, so as to obtain data on customers with high-value insured products and customers with low-value insured products;

Perform customer intersection calculation based on the data of customers with a high proportion of long-term products and the data of customers with a high number of insurance premiums, so as to obtain the data of the first most loyal customers;

Perform customer intersection calculation based on the data of customers with a high proportion of short-term products and the data of customers with a high number of insurance premiums, thereby obtaining the data of the second most loyal customers;

Perform customer intersection calculation based on the data of customers with the same proportion and the data of customers with high insurance amount, so as to obtain the data of the third most loyal customers;

Perform customer intersection calculation based on the monotonous product customer data and the high-value insurance customer data to obtain the fourth most loyal customer data;

Perform customer intersection operations based on the monotonous product customer data and the low-amount insured customer data to obtain low-loyalty customer data;

The first high-loyalty customer data, the second high-loyalty customer data, the third high-loyalty customer data and the fourth high-loyalty customer data are merged to obtain high-loyalty customer data, and the high-loyalty customer data and the low-loyalty customer data are merged to obtain customer loyalty data.

These customers in the customer data with a high proportion of short-term products of the present invention may prefer short-term products, may need more frequent services or be more susceptible to market changes. These customers in the customer data with a high proportion of long-term products may prefer long-term products, may need more stable investments or pay more attention to long-term protection. These customers in the customer data of monotonous products may have a higher preference for long-term or short-term products, may have more specific needs, and need to focus on their product satisfaction and demand changes. These customers in the customer data with the same proportion may have similar proportions in different product types, and may have certain needs and preferences for different types of products. These customers in the customer data with a high number of insured persons may have purchased multiple products, may be high net worth customers or customers interested in multiple protections. These customers in the customer data with a low number of insured persons may have purchased fewer products, may be new customers, small investors or customers who do not have a deep understanding of the products. By performing intersection operations on customer data under different conditions, such as the intersection of a high proportion of long-term products and a high number of insured customers, customer groups that meet multiple conditions can be found, and these customers may have higher loyalty and higher investment needs. By identifying the intersection of monotonous product customers and low-insurance customers, we can find customer groups that may need extra attention and guidance. These customers may not have a deep understanding of the product or are more sensitive to market fluctuations. Combining customer data at different loyalty levels can form a comprehensive customer loyalty data, which is helpful for formulating marketing strategies and service plans for customers at different loyalty levels.

Optionally, step S36 is specifically:

Step S361: extracting features from the insurance product data to be evaluated, thereby obtaining the term data and premium data of the product to be evaluated;

Step S362: extracting the customer's historical insurance product data from the insurance product history data according to the customer loyalty data, thereby obtaining the customer's historical insurance product data;

Step S363: Calculate the product term proportion of the customer's historical insurance product data, thereby obtaining the customer's historical product term proportion data, and calculate the term coverage of the customer's historical product term proportion data according to the product term data to be evaluated, thereby obtaining the customer's term coverage data;

Step S364: Calculate the product premium ratio of the customer's historical insurance product data, thereby obtaining the customer's historical product premium ratio data, and calculate the average premium of the customer's historical insurance product data, thereby obtaining the customer's average premium data;

Step S365: Calculate the premium dispersion of the product to be evaluated and the average premium data of the customer, thereby obtaining premium dispersion data; calculate the premium coverage of the customer's historical product premium ratio data based on the premium dispersion data, thereby obtaining the customer's premium coverage data;

Step S366: Evaluate the product coverage of the customer's historical insurance product data based on the customer's premium coverage data and the customer's term coverage data, thereby obtaining coverage evaluation data.

By extracting the characteristics of the insurance product to be evaluated, the present invention can understand the term and premium distribution of the product to be evaluated, and provide basic data for subsequent analysis. By using customer loyalty data, the customer's historical insurance products can be screened out, and the different insurance products purchased by the customer in the past can be understood, providing a data basis for subsequent analysis. The customer's historical product term proportion data and term coverage data are helpful to evaluate the term distribution of the customer's historical insurance products, and determine whether the term of the product to be evaluated matches the customer's historical policy term, and provide the customer's preference for products with different terms. The customer's historical product premium proportion and average premium data can be used to understand the premium distribution of products purchased by the customer in the past, as well as the customer's average premium level. This is very helpful for evaluating the customer's ability to pay and premium preferences. By analyzing the customer's historical product premium proportion and average premium data, the customer's ability to pay and risk tolerance can be evaluated. This helps insurance companies better manage risks, formulate reasonable pricing strategies, and ensure the sustainability and profitability of insurance products. By evaluating the customer's historical insurance product data and premium coverage data, the customer's insurance needs and preferences can be better understood, and more personalized and accurate services can be provided to customers. This helps to establish long-term and stable customer relationships and improve customer satisfaction and loyalty. By evaluating the customer's historical product coverage, we can evaluate the product's performance in meeting customer needs and provide a reference for product promotion and marketing. This helps optimize sales strategies and performance evaluation systems, and improve sales efficiency and the success rate of product promotion.

Optionally, step S4 is specifically:

Step S41: extracting payment period features from the insurance product data to be evaluated, thereby obtaining product payment period data;

Step S42: Calculate the product default interaction frequency according to the product payment period data, thereby obtaining the default interaction frequency data of the product to be tested;

Step S43: extracting product payment period features and interaction frequency features from historical insurance product data, thereby obtaining historical product payment period data and historical product actual interaction frequency data;

Step S44: Conduct customer interaction experience evaluation on historical product payment period data, historical product actual interaction frequency data, and the default interaction frequency data of the product to be tested, thereby obtaining interaction experience evaluation data.

By extracting the payment period characteristics of the product, the present invention can understand the payment cycle and method of the product, and then provide basic data for subsequent interaction frequency calculation and customer interaction experience evaluation. It is helpful to evaluate the flexibility and customizability of the product, understand the customer's preferences in payment, and thus provide a reference for product customization and promotion. Calculating the default interaction frequency of the product to be tested helps to understand the degree of interaction between the customer and the product, and then provide a reference for product design and sales strategy. The frequency of use and customer participation of the product can be evaluated based on the interaction frequency data, providing guidance for product improvement and market positioning. By extracting the payment period characteristics and actual interaction frequency data of historical products, the characteristics of the product to be tested can be compared to understand the advantages and disadvantages of the product and its market competitiveness. It is helpful to discover the performance of historical products in terms of payment cycle and interaction frequency, and provide reference opinions for product improvement and optimization. Through customer interaction experience evaluation, customers' attitudes and preferences for product payment period and interaction frequency can be understood, thereby providing guidance for product improvement and market promotion. Customer satisfaction and experience during actual use of the product can be evaluated, providing improvement suggestions for product design and service improvement.

Optionally, step S44 is specifically:

Step S441: Calculate the default interaction frequency according to the historical product payment period data, thereby obtaining the historical product default interaction frequency data;

Step S442: performing a difference calculation on the historical product default interaction frequency data and the historical product actual interaction frequency data, thereby obtaining the historical product interaction difference data;

Step S443: extracting difference product data from historical insurance product data according to historical product interaction difference data, thereby obtaining interaction difference product data;

Step S444: performing similarity calculation on the interactive difference product data and the insurance product data to be evaluated, thereby obtaining product similarity data, and extracting high-similarity product data from the interactive difference product data according to the product similarity data, thereby obtaining high-similarity product data;

Step S445: Predicting the actual interaction frequency of the insurance product data to be evaluated based on the high-similarity product data, thereby obtaining actual interaction frequency prediction data of the product to be evaluated;

Step S446: Calculate the default interaction ratio based on the default interaction frequency data of the product to be tested and the actual interaction frequency prediction data of the product to be tested, so as to obtain interaction experience evaluation data.

The present invention calculates the default interaction frequency through the payment period information in the historical data, understands the average customer interaction frequency of past products, and provides benchmark data for comparison and analysis. It is helpful to discover the general trend of products in past customer interactions and provide a reference for product design and market promotion. By calculating the difference between the default interaction frequency and the actual interaction frequency, the changes and trends of historical products in customer interaction can be identified, which helps to understand the actual mode of customer use of products. Interaction difference data for subsequent analysis is provided, which helps to discover the actual needs and expectations of customers for products. By extracting the interaction difference data of historical products, a set of product data reflecting the changes in customer interaction can be obtained, which helps to deeply understand the evolution process of products and the dynamics of customer interaction. Through similarity calculation, the similarity between historical products and products to be evaluated can be identified, providing a reference for product positioning, improvement and market promotion. Extracting high-similarity product data helps to find historical products with similar customer interaction patterns to the products to be evaluated, providing a basis for subsequent prediction and evaluation. Using the historical interaction data of high-similarity products, the actual customer interaction frequency of the products to be tested can be predicted, providing a basis for the expected interaction after the product is launched. It provides possible customer interaction patterns for the products to be tested in the future, which helps to formulate product design and market strategies. By calculating the ratio of default interaction frequency to actual interaction frequency, you can evaluate the customer interaction experience of the product and find out whether customers interact with the product as frequently as expected. This helps identify the product's advantages and improvement points in customer experience and provides a reference for product promotion and service improvement.

Optionally, this specification also provides an insurance product evaluation system for executing the insurance product evaluation method as described above, the insurance product evaluation system comprising:

The risk assessment model building module is used to obtain historical data of insurance products and conduct claims risk assessment on the historical data of insurance products, thereby obtaining historical claims risk data; and building a claims risk assessment model based on the historical claims risk data;

The claim risk assessment module is used to obtain the insurance product data to be assessed, and to conduct claim risk assessment on the insurance product data to be assessed through the claim risk assessment model, thereby obtaining claim risk assessment data;

The coverage evaluation module is used to extract customer features from historical insurance product data to obtain historical customer data, and to classify historical customer data by customer loyalty to obtain customer loyalty data; and to conduct coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data to obtain coverage evaluation data;

The interactive experience evaluation module is used to calculate the product default interaction frequency of the insurance product data to be evaluated, thereby obtaining the default interaction frequency data of the product to be tested, and to evaluate the customer interactive experience of the default interaction frequency data of the product to be tested and the historical data of the insurance product, thereby obtaining the interactive experience evaluation data;

The comprehensive product evaluation module is used to conduct a comprehensive evaluation of insurance products based on claims risk evaluation data, coverage evaluation data, and interactive experience evaluation data, so as to obtain an insurance product evaluation report.

The insurance product evaluation system of the present invention can implement any insurance product evaluation method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the insurance product evaluation method. The internal modules of the system cooperate with each other, thereby improving the accuracy of the insurance product evaluation results.

BRIEF DESCRIPTION OF THE DRAWINGS

Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

FIG1 is a schematic diagram of the steps of the insurance product evaluation method of the present invention;

FIG2 is a schematic diagram of a detailed step flow chart of step S1 in the present invention;

FIG. 3 is a schematic diagram of a detailed flow chart of step S3 in the present invention.

The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings.

Detailed ways

The following is a clear and complete description of the technical method of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and/or processor methods and/or microcontroller methods.

It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly a second unit may be referred to as a first unit. The term "and/or" used herein includes any and all combinations of one or more of the listed associated items.

To achieve the above object, please refer to Figures 1 to 3. The present invention provides an insurance product evaluation method, comprising the following steps:

Step S1: Obtain historical data of insurance products, and conduct claims risk assessment on the historical data of insurance products, thereby obtaining historical claims risk data; and construct a claims risk assessment model based on the historical claims risk data;

In this embodiment, historical data of past insurance products are collected, including information of the insured, details of the policy, and claims records. Through careful data analysis, statistical methods and machine learning techniques are used to evaluate the claims risk of historical data, and detailed historical claims risk data are generated. Based on these data, a claims risk evaluation model is established, and a model is constructed using algorithms such as logistic regression or decision trees in combination with variables such as the age, occupation, and health status of the insured, so as to quantify the claims risk.

Step S2: Obtain the insurance product data to be evaluated, and perform a claims risk evaluation on the insurance product data to be evaluated using a claims risk evaluation model, thereby obtaining claims risk evaluation data;

In this embodiment, the insurance product data to be evaluated is obtained, including product features, insured object information, etc. These data are evaluated through the previously constructed claim risk evaluation model to calculate the claim risk of the product to be tested, that is, the claim risk evaluation data. This can help understand the potential risk level of the product and provide a scientific basis for formulating product strategies.

Step S3: extracting customer features from historical data of insurance products to obtain historical customer data, and classifying historical customer data by customer loyalty to obtain customer loyalty data; conducting coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data to obtain coverage evaluation data;

In this embodiment, customer characteristics are extracted from the historical data of insurance products, including purchase frequency, policy renewal status, etc. By clustering or classifying the historical customer data, the loyalty data of different customer groups is obtained. Based on these data, the potential coverage of the product to be evaluated is evaluated to understand the market penetration of the product in different customer groups.

Step S4: Calculate the product default interaction frequency for the insurance product data to be evaluated, thereby obtaining the default interaction frequency data for the product to be tested, and perform customer interaction experience evaluation on the default interaction frequency data for the product to be tested and the insurance product historical data, thereby obtaining interaction experience evaluation data;

In this embodiment, the data of the product to be evaluated is used to calculate the default interaction frequency, that is, the frequency of customer interaction with the product without a specific claim event. Combined with historical data, the product's interactive experience is evaluated, including indicators such as customer satisfaction and feedback rate, to form detailed interactive experience evaluation data.

Step S5: Perform a comprehensive evaluation of the insurance product based on the claims risk evaluation data, coverage evaluation data, and interactive experience evaluation data to obtain an insurance product evaluation report.

In this embodiment, the evaluation data of the three aspects of claim risk, coverage and interactive experience are comprehensively considered, and weighed and comprehensively evaluated. The final insurance product evaluation report is generated by using appropriate scoring models such as weighted total score model and hierarchical analysis method, providing a comprehensive evaluation of the product and providing detailed suggestions and data support for product improvement, market positioning and promotion strategy.

The present invention constructs a claims risk assessment model by analyzing the historical claims data of insurance products. Such a model can identify and quantify different types of risks, providing an important basis for subsequent insurance product evaluation. By understanding the historical claims data, the potential claims risks in the future can be better predicted. By performing claims risk assessment on the insurance product data to be evaluated, the risk level of the product can be quantified. This helps insurance companies better understand the risk characteristics of each product, so as to formulate pricing strategies and claims policies more effectively. By extracting customer characteristics and classifying loyalty from historical customer data, customers' behaviors and preferences can be better understood. Such an assessment can help insurance companies better understand their existing customer groups, design more attractive insurance products and services, and improve customer loyalty. By calculating the default interaction frequency of the product and the customer interaction experience assessment, the interactivity and customer friendliness of the insurance product can be evaluated. This helps insurance companies design more attractive and easy-to-use products, improve customer experience, and thus enhance customer satisfaction and loyalty. By comprehensively considering the claims risk assessment, coverage assessment, and interaction experience assessment data, a comprehensive evaluation report of the insurance product can be obtained. Such a comprehensive assessment can help insurance companies fully understand the advantages and room for improvement of the product, and provide an important reference for business decision-making.

Optionally, step S1 specifically includes:

Step S11: Obtaining historical data of insurance products;

In this embodiment, the insurance product historical data is obtained through the internal database of the insurance company, data provided by partners or public channels. The insurance product historical data contains detailed insurance product information, including but not limited to policyholder information, premiums, claims records, policy terms, etc.

Step S12: extracting the term feature of the historical data of insurance products, thereby obtaining historical long-term insurance product data and historical short-term insurance product data;

In this embodiment, a data processing tool is used to extract the term feature of the historical data of insurance products. This includes extracting information such as the policy term and payment term from each insurance product. For example, key features such as the insurance term and payment term start and end dates of each insurance product, as well as the policy duration, can be calculated.

Step S13: conducting a long-term product claim risk assessment on historical long-term insurance product data, thereby obtaining long-term product claim risk data;

In this embodiment, historical data of long-term insurance products are used to perform claims risk assessment. This may include establishing a mathematical model, such as a risk model based on historical claims frequency and claims amount. Through statistical analysis and machine learning methods, the claims risk level of long-term products is determined and corresponding assessment data is generated.

Step S14: conducting a short-term product claim risk assessment on historical short-term insurance product data, thereby obtaining short-term product claim risk data;

In this embodiment, the claim risk assessment is performed on the historical data of short-term insurance products. Using a method similar to that of long-term products, a model can be established to assess the claim risk of short-term products. This may involve different feature extraction and model parameters to adapt to the particularity of short-term products.

Step S15: merging the long-term product claim risk data and the short-term product claim risk data to obtain historical claim risk data;

In this embodiment, the long-term product claim risk data and the short-term product claim risk data are merged to ensure that the data format is consistent. Data cleaning and conversion may be required so that the information of these two types of products can be seamlessly integrated in subsequent steps.

Step S16: Construct a claims risk assessment model based on historical claims risk data.

In this embodiment, a claims risk assessment model is constructed based on the integrated historical claims risk data. The historical claims risk data can be modeled using methods including statistical modeling and machine learning algorithms. By training the model, the claims risk of future insurance products can be predicted.

By acquiring historical data of insurance products, the present invention can establish a comprehensive data set, which includes past insurance product information, policy details, claims records, etc. This provides basic data for subsequent steps, so that the model can take into account the impact of historical information in the claims risk assessment. Through term feature extraction, the insurance product data can be divided into long-term and short-term, and then the risks of these two types of products can be analyzed in a targeted manner. This helps to more accurately assess the differences in claims risks between long-term and short-term insurance products. By performing claims risk assessment on historical long-term insurance product data, the potential risks of long-term products can be identified and quantified. This helps insurance companies better understand and manage the risks of long-term products and adopt corresponding insurance strategies. Similar to long-term products, performing claims risk assessment on historical short-term insurance product data helps to identify and quantify the risks of short-term products. This enables insurance companies to better understand and respond to the claims risks that short-term products may face. By merging the claims risk data of long-term and short-term products, a more comprehensive historical claims risk data set can be created. This helps to comprehensively consider the risk factors of long-term and short-term products and provide richer information for building a comprehensive claims risk assessment model. A claims risk assessment model is constructed, which is based on historical claims risk data and can predict the claims risks of future insurance products. This provides insurers with a powerful tool to help them price products more accurately, develop risk management strategies, and improve overall business efficiency.

Optionally, step S13 is specifically:

Step S131: extracting features from historical long-term insurance product data, thereby obtaining historical long-term claims data and historical long-term premium payment data;

In this embodiment, by extracting features from historical long-term insurance product data, key information can be obtained, including claims records and premium payments. For claims data, information such as the number of claims and the amount of claims for each policy can be extracted. For premium payment data, information such as the payment frequency and the amount of premiums for each policy can be extracted. In this way, a historical data set of claims and premium payments for long-term insurance products can be established.

Step S132: Calculate the claim probability of the same product based on the historical long-term claim data to obtain product claim probability data, and perform statistical analysis on the product claim probability data to obtain the product claim risk threshold;

In this embodiment, the historical long-term claims data can be used to calculate the claim probability of the same product. By statistically analyzing these data, the probability distribution of product claims can be determined, and a threshold for claim risk can be set based on this. This threshold can help classify long-term products into high-risk and low-risk categories.

Step S133: classifying the historical long-term claims data according to the product claims risk threshold, thereby obtaining high claims risk long-term product data and low claims risk long-term product data;

In this embodiment, according to the set product claim risk threshold, the historical long-term claim data can be classified into two categories: high claim risk and low claim risk. In this way, high claim risk long-term product data and low claim risk long-term product data can be obtained, providing a basis for subsequent risk management and product strategy formulation.

Step S134: Counting the number of missed payments by customers based on historical long-term premium payment data, thereby obtaining missed payment data of high-frequency customers and missed payment data of low-frequency customers;

In this embodiment, the historical long-term premium payment data can be used to count the number of missed payments for each customer. In this way, it is possible to distinguish between high-frequency missed payments and low-frequency missed payments. This information can help assess the customer's payment credit level, thereby better controlling risks.

Step S135: classifying the historical long-term insurance product data by customer risk according to the high-frequency customer missed payment data and the low-frequency customer missed payment data, thereby obtaining high-customer risk long-term product data and low-customer risk long-term product data;

In this embodiment, the customer missed payment data is combined with the historical long-term insurance product data to classify customer risks. In this way, high customer risk long-term product data and low customer risk long-term product data can be obtained, which is helpful for formulating risk management strategies for different customer groups.

Step S136: performing an intersection operation on the long-term product data with high claim risk and the long-term product data with high customer risk, thereby obtaining the long-term product claim high risk data; performing an intersection operation on the long-term product data with low claim risk and the long-term product data with low customer risk, thereby obtaining the long-term product claim low risk data;

In this embodiment, by performing an intersection operation on the long-term product data of high claims risk and high customer risk, the long-term product claims high risk data is obtained; similarly, by performing an intersection operation on the long-term product data of low claims risk and low customer risk, the long-term product claims low risk data is obtained.

Step S137: Merge the long-term product claim low-risk data and the long-term product claim high-risk data to obtain the long-term product claim risk data.

In this embodiment, the long-term product claim low risk data and the long-term product claim high risk data are merged to obtain a complete long-term product claim risk data set. This data set can be used in decision-making processes such as risk management, product design and pricing.

The present invention can extract useful features from historical data through feature extraction for subsequent risk assessment and classification. The calculation of claim probability and risk threshold can help insurance companies better understand the risk level of products and provide a basis for risk classification. Classifying products according to risk level can help insurance companies formulate risk management strategies more targeted and improve the effectiveness of risk control. Statistics on the number of missed payments by customers can help insurance companies understand customers' payment situations, identify high-risk customers, and take corresponding measures to reduce risks. Classifying customers according to risk levels can help insurance companies better understand the risk characteristics of customer groups and carry out customer management and risk control in a targeted manner. By performing intersection operations on claim risk and customer risk, high-risk products and customers can be identified more accurately, providing more sophisticated data support for risk management. Merging low-risk and high-risk data can obtain a complete risk data set, providing insurance companies with a more comprehensive basis for risk assessment and supporting decisions such as insurance product design and pricing.

Optionally, step S14 is specifically:

Step S141: extracting features from historical short-term insurance product data to obtain historical short-term claims data and customer age data;

In this embodiment, feature extraction is performed on historical short-term insurance product data, including customer ID, policy start and end dates, insured amount, premium, etc. At the same time, the customer's age information is extracted from these data.

Step S142: performing claim frequency statistics on historical short-term claim data, thereby obtaining high-frequency short-term claim data and low-frequency short-term claim data;

In this embodiment, the historical short-term claims data is subjected to claims frequency statistics, which may include counting the number of claims for each customer to identify high-frequency and low-frequency claims customers.

Step S143: classify and calculate the customer age data, so as to obtain the age data of older customers and the age data of younger customers;

In this embodiment, the customer age data is classified and calculated according to the common sense of growth age, for example, the customer age is divided into two groups: the old group and the young group.

Step S144: extracting age-restricted insurance products from historical short-term insurance product data based on the age data of senior customers and the age data of junior customers, thereby obtaining short-term age-restricted insurance product data; performing intersection operations on low-frequency short-term claims data and short-term age-restricted insurance product data, thereby obtaining short-term product claims low-risk data;

In this embodiment, the corresponding age-restricted insurance products are extracted from the historical short-term insurance product data based on the age data of senior customers and the age data of junior customers. Then, the low-frequency short-term claims data and the age-restricted insurance product data are intersected, and the product data that appears in both data is extracted to obtain low-risk short-term product data, that is, short-term product claims low-risk data.

Step S145: performing customer association on the high-frequency short-term claims data and the age data of senior customers, thereby obtaining first customer association data, and extracting high-risk senior customer product data from the historical short-term insurance product data based on the first customer association data, thereby obtaining high-risk data on senior short-term product claims;

In this embodiment, the customer ID in the high-frequency short-term claims data is associated with the customer ID in the age data of the elderly customer to obtain the first customer-related data. The customer ID in the first customer-related data is used to filter the data related to these customers in the historical short-term insurance product data. According to business rules and risk standards, high-risk elderly customers are identified, such as customers whose age exceeds a certain threshold and who have a high-frequency claims record. Short-term insurance product data related to these customers, i.e., high-risk data for elderly short-term product claims, are extracted.

Step S146: performing customer association on the high-frequency short-term claims data and the age data of young customers, thereby obtaining second customer association data, and extracting high-risk young customer product data from the historical short-term insurance product data based on the second customer association data, thereby obtaining high-risk data on claims of young short-term products;

In this embodiment, the customer ID in the high-frequency short-term claims data is associated with the customer ID in the low-age customer age data to obtain the second customer-related data. The customer ID in the second customer-related data is used to filter the data related to these customers in the historical short-term insurance product data. For example, products with low-age participation and high-frequency claims records are extracted. The relevant short-term insurance product data, i.e., the high-risk data of claims for low-age short-term products, is extracted.

Step S147: Merge the low-risk data on short-term product claims, the high-risk data on senior short-term product claims, and the high-risk data on junior short-term product claims to obtain short-term product claim risk data.

In this embodiment, the low-risk data of short-term product claims, the high-risk data of senior short-term product claims, and the high-risk data of junior short-term product claims are merged. Ensure that consistent data formats and field identifiers are used during the merging. The merging operation can be performed using database operations or data processing tools.

Through feature extraction, the present invention can understand the customer's claims situation and age distribution, and provide a data basis for subsequent risk analysis and customer association. Claim frequency statistics can help identify products with frequent claims, and these data can be regarded as potential high-risk factors. Through the classification of age data, risk analysis and customized insurance product design can be performed for customers of different age groups. By limiting the age range and claim frequency, safer insurance products and customers can be extracted to reduce the risk of claims. Through association analysis, the common characteristics of high-risk customer groups can be discovered, and risk management and product design can be carried out in a targeted manner. By merging data at different risk levels, comprehensive claims risk data can be obtained, providing support for insurance companies to formulate more effective risk management strategies.

Optionally, step S3 specifically includes:

Step S31: extracting customer features from historical insurance product data to obtain historical customer data;

In this embodiment, customer related information is extracted from the insurance product historical data, such as customer ID, gender, age, occupation, family situation, etc. Appropriate data processing tools and techniques, such as SQL query or Python programming, are used to extract the required customer feature information from the insurance product historical data.

Step S32: extracting the insured product data from the historical customer data, thereby obtaining the customer's insured product data;

In this embodiment, historical customer data is used to extract the product information of the customer's insurance, including product ID, product type, insurance time, insurance amount, etc. This can be done by querying the policy information in the historical data, or by connecting the customer data and the policy data.

Step S33: performing insurance period statistics on the customer insurance product data, thereby obtaining long-term customer insurance product data and short-term customer insurance product data, and calculating the insurance product ratio for the long-term customer insurance product data and short-term customer insurance product data, thereby obtaining customer insurance product ratio data;

In this embodiment, the customer insurance product data is statistically analyzed to distinguish long-term customers from short-term customers. For each customer, the insurance period of the insurance product is calculated, and those whose insurance period exceeds a certain threshold (such as one year) are defined as long-term customers, otherwise they are defined as short-term customers. The insurance product data of long-term customers and short-term customers are calculated, and the proportion of long-term customers and short-term customers is calculated.

Step S34: Count the number of insured products based on the customer insurance data, thereby obtaining the number of insured products for the customer;

In this embodiment, the customer's insured product data is counted to calculate the number of products insured by each customer, which can be the total number or the number of different types of products.

Step S35: dividing the customer's insured product ratio data and the number of insured products by customer into customer loyalty categories, thereby obtaining customer loyalty data;

In this embodiment, customer loyalty classification rules are formulated based on the customer's insured product ratio data and the number of customer's insured products. Customers can be divided into high, medium, and low loyalty levels according to business needs, or classification rules can be formulated based on analysis of specific data.

Step S36: Conduct coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data, thereby obtaining coverage evaluation data.

In this embodiment, customer loyalty data is used to evaluate the coverage of the insurance product data to be evaluated. According to the characteristics of different levels of customer loyalty, the coverage and potential market of each insurance product in different loyalty customer groups are evaluated. Statistical analysis or machine learning models can be used for evaluation and prediction.

By extracting customer characteristics, the present invention can understand the personal and insurance-related information of historical customers and provide a data basis for subsequent analysis. By extracting customer insurance product data, the customer's insurance purchasing behavior can be understood to form a customer-product association. Understanding the insurance behavior of long-term and short-term customers and calculating the proportion of insured products helps to understand the customer's insurance purchasing preferences and provide guidance for product design and sales strategies. By counting the number of insured products, the customer's insurance portfolio can be understood, providing a reference for product recommendations and pricing. Dividing customers into different loyalty levels can help companies identify high-loyalty customers and low-loyalty customers, thereby formulating corresponding customer relationship management strategies. By evaluating the coverage of the insurance products to be evaluated, the matching degree between the products and customers with different loyalty can be understood, providing data support for product improvement and positioning.

Optionally, the customer loyalty classification in step S35 is specifically as follows:

Conduct product ratio statistical analysis on the customer insurance product ratio data, so as to obtain customer data with a high proportion of short-term products, customer data with a high proportion of long-term products, customer data with monotonous products, and customer data with the same proportion;

In this embodiment, the definition of short-term products is determined based on the insurance period of insurance products in historical data (such as an insurance period of less than one year). Analyze the customer's insurance product data and calculate the proportion of short-term products for each customer. Identify customers whose proportion is higher than a certain threshold (such as 50%) and classify them as customers with a high proportion of short-term products. Similar to short-term products, the definition of long-term products is determined based on the insurance period of the insured products (such as an insurance period of more than one year). Analyze the customer's insurance product data and calculate the proportion of long-term products for each customer. Identify customers whose proportion is higher than a certain threshold and classify them as customers with a high proportion of long-term products. A monotonous product is a product in which a customer only purchases one type (long-term or short-term). Analyze the customer's insurance product data, identify customers who only insure one product, and classify them as monotonous product customers. Identify customer groups with the same proportion of insured products, that is, the proportion of insured products of multiple customers is similar and there is no significant difference.

Conduct product quantity statistics analysis on the number of insured products purchased by customers, so as to obtain data on customers with high-value insured products and customers with low-value insured products;

In this embodiment, the number of insured products for each customer is counted. Customers whose insured product number is higher than a certain threshold (such as 3 products) are identified and classified as high-value insured customers. The identification of high-value insured customers is similar to high-value insured quantity, but a lower threshold (such as 1 or 2 products) is set.

Perform customer intersection calculation based on the data of customers with a high proportion of long-term products and the data of customers with a high number of insurance premiums, so as to obtain the data of the first most loyal customers;

In this embodiment, the intersection operation is performed on the customer data with a high proportion of long-term products and the customer data with a high number of insured persons, and the customer data with the same two types of data are counted to obtain the first high-loyalty customer data. If customers purchase a large proportion of long-term products, it means that they have confidence in the long-term and stable relationship with the insurance company, and purchasing multiple products may mean that they are more inclined to fully cover their protection needs. These are all manifestations of loyal customers, so customers with both a high proportion and a high number of insured persons are defined as high-loyalty customers.

Perform customer intersection calculation based on the data of customers with a high proportion of short-term products and the data of customers with a high number of insurance premiums, thereby obtaining the data of the second most loyal customers;

In this embodiment, the intersection operation is performed on the customer data with a high proportion of short-term products and the customer data with a high number of insured persons, and the customer data with the same two types of data are counted to obtain the second most loyal customer data. The fact that customers purchase a large proportion of short-term products means that they recognize the company's products, and purchasing multiple products may mean that they prefer to fully cover their protection needs. These are all manifestations of loyal customers, so customers with both a high proportion and a high number of insured persons are defined as high-loyalty customers.

Perform customer intersection calculation based on the data of customers with the same proportion and the data of customers with high insurance amount, so as to obtain the data of the third most loyal customers;

In this embodiment, the intersection operation is performed on the customer data with the same proportion and the customer data with a high amount of insurance participation, and the customer data with the same two types of data are counted to obtain the third highest loyalty customer data. The long-term products and short-term products purchased by customers are in the same proportion, which means that these customers have confidence in the long-term and stable relationship with the insurance company and also recognize the insurance company's products. Purchasing multiple products may mean that they are more inclined to fully cover their protection needs. These are all manifestations of loyal customers, so customers with the same proportion and a high amount of insurance participation are defined as high-loyalty customers.

Perform customer intersection calculation based on the monotonous product customer data and the high-value insurance customer data to obtain the fourth most loyal customer data;

In this embodiment, the customer data of monotonous products and the customer data of high-value insured persons are intersected, and the customer data of the two types of data are counted to obtain the fourth highest loyalty customer data. The purchase of long-term products or short-term products monotonically represents the customer's trust in the company's products. Long-term product buyers may be more inclined to establish long-term cooperative relationships and show higher loyalty. Customers with high-value insured persons show a comprehensive dependence on the company's products, which means that they have chosen the company's products in multiple protection areas, which also reflects a certain degree of loyalty. The customer data of high-value insured persons and the customer data of monotonous products are intersected to screen out customers who show loyalty not only in terms of product quantity, but also in terms of product type. This multi-faceted loyalty performance more comprehensively reflects the customer's loyalty level. Considering the customer data of monotonous products or the customer data of high-value insured persons alone may result in certain misjudgments, and it is impossible to fully and accurately evaluate the customer's loyalty. Through the intersection operation, the indicators of the two aspects can complement each other, reduce the risk of misjudgment, and more accurately determine high-loyalty customers.

Perform customer intersection operations based on the monotonous product customer data and the low-amount insured customer data to obtain low-loyalty customer data;

In this embodiment, the intersection operation is performed on the monotonous product customer data and the low-amount insurance customer data, and the customer data that appear in the same two data are counted to obtain low-loyalty customer data, which is relative to these high-loyalty customers. Its characteristics may be a low proportion of purchased products, a small number of products, or a lack of overlap between different loyalty indicators. These customers may not have a high dependence on the company's products or services.

The first high-loyalty customer data, the second high-loyalty customer data, the third high-loyalty customer data and the fourth high-loyalty customer data are merged to obtain high-loyalty customer data, and the high-loyalty customer data and the low-loyalty customer data are merged to obtain customer loyalty data.

In this embodiment, the first high-loyalty customer data, the second high-loyalty customer data, the third high-loyalty customer data and the fourth high-loyalty customer data are combined into one data set. The low-loyalty customer data and the high-loyalty customer data are combined into one data set to obtain complete customer loyalty data.

These customers in the customer data with a high proportion of short-term products of the present invention may prefer short-term products, may need more frequent services or be more susceptible to market changes. These customers in the customer data with a high proportion of long-term products may prefer long-term products, may need more stable investments or pay more attention to long-term protection. These customers in the customer data of monotonous products may have a higher preference for long-term or short-term products, may have more specific needs, and need to focus on their product satisfaction and demand changes. These customers in the customer data with the same proportion may have similar proportions in different product types, and may have certain needs and preferences for different types of products. These customers in the customer data with a high number of insured persons may have purchased multiple products, may be high net worth customers or customers interested in multiple protections. These customers in the customer data with a low number of insured persons may have purchased fewer products, may be new customers, small investors or customers who do not have a deep understanding of the products. By performing intersection operations on customer data under different conditions, such as the intersection of a high proportion of long-term products and a high number of insured customers, customer groups that meet multiple conditions can be found, and these customers may have higher loyalty and higher investment needs. By identifying the intersection of monotonous product customers and low-insurance customers, we can find customer groups that may need extra attention and guidance. These customers may not have a deep understanding of the product or are more sensitive to market fluctuations. Combining customer data at different loyalty levels can form a comprehensive customer loyalty data, which is helpful for formulating marketing strategies and service plans for customers at different loyalty levels.

Optionally, step S36 is specifically:

Step S361: extracting features from the insurance product data to be evaluated, thereby obtaining the term data and premium data of the product to be evaluated;

In this embodiment, the feature extraction of the insurance product data to be evaluated can be performed by using data mining technology, including but not limited to extracting product term data and premium data. For example, by analyzing the policy information, the term of the insurance product and the corresponding premium information can be extracted.

Step S362: extracting the customer's historical insurance product data from the insurance product history data according to the customer loyalty data, thereby obtaining the customer's historical insurance product data;

In this embodiment, based on the customer loyalty data, the customer's historical insurance product data is extracted from the insurance product history data. This can be done by associating the customer ID and screening its historical insurance records to obtain a list of insurance products held by the customer.

Step S363: Calculate the product term proportion of the customer's historical insurance product data, thereby obtaining the customer's historical product term proportion data, and calculate the term coverage of the customer's historical product term proportion data according to the product term data to be evaluated, thereby obtaining the customer's term coverage data;

In this embodiment, the product term ratio of the customer's historical insurance product data is calculated. By counting the ratio of each product term in the customer's history, the customer's historical product term ratio data is obtained. The term coverage is calculated using the term data of the product to be evaluated, reflecting the matching degree between the customer's historical insurance product term and the product to be evaluated.

Step S364: Calculate the product premium ratio of the customer's historical insurance product data, thereby obtaining the customer's historical product premium ratio data, and calculate the average premium of the customer's historical insurance product data, thereby obtaining the customer's average premium data;

In this embodiment, the product premium ratio is calculated for the customer's historical insurance product data, and the average premium is calculated. By counting the ratio of each product premium in the customer's history, the customer's historical product premium ratio data is obtained. At the same time, the average premium is calculated as the customer's typical premium level.

Step S365: Calculate the premium dispersion of the product to be evaluated and the average premium data of the customer, thereby obtaining premium dispersion data; calculate the premium coverage of the customer's historical product premium ratio data based on the premium dispersion data, thereby obtaining the customer's premium coverage data;

In this embodiment, the premium dispersion is calculated for the premium data of the product to be evaluated and the average premium data of the customer. This can be done by using statistical methods such as standard deviation to reflect the degree of dispersion of the premium. Furthermore, based on the premium dispersion data, the premium coverage of the customer's historical product premium ratio is calculated to reveal the comprehensiveness of the premium distribution in the customer's historical products.

Step S366: Evaluate the product coverage of the customer's historical insurance product data based on the customer's premium coverage data and the customer's term coverage data, thereby obtaining coverage evaluation data.

In this embodiment, the customer's premium coverage data and term coverage data are combined to evaluate the product coverage of the customer's historical insurance products. This can be done by comprehensively considering the term and premium coverage, evaluating the overall performance of the customer's historical insurance products in terms of matching with the product to be evaluated, and forming product coverage evaluation data.

By extracting the characteristics of the insurance product to be evaluated, the present invention can understand the term and premium distribution of the product to be evaluated, and provide basic data for subsequent analysis. By using customer loyalty data, the customer's historical insurance products can be screened out, and the different insurance products purchased by the customer in the past can be understood, providing a data basis for subsequent analysis. The customer's historical product term proportion data and term coverage data are helpful to evaluate the term distribution of the customer's historical insurance products, and determine whether the term of the product to be evaluated matches the customer's historical policy term, and provide the customer's preference for products with different terms. The customer's historical product premium proportion and average premium data can be used to understand the premium distribution of products purchased by the customer in the past, as well as the customer's average premium level. This is very helpful for evaluating the customer's ability to pay and premium preferences. By analyzing the customer's historical product premium proportion and average premium data, the customer's ability to pay and risk tolerance can be evaluated. This helps insurance companies better manage risks, formulate reasonable pricing strategies, and ensure the sustainability and profitability of insurance products. By evaluating the customer's historical insurance product data and premium coverage data, the customer's insurance needs and preferences can be better understood, and more personalized and accurate services can be provided to customers. This helps to establish long-term and stable customer relationships and improve customer satisfaction and loyalty. By evaluating the customer's historical product coverage, we can evaluate the product's performance in meeting customer needs and provide a reference for product promotion and marketing. This helps optimize sales strategies and performance evaluation systems, and improve sales efficiency and the success rate of product promotion.

Optionally, step S4 is specifically:

Step S41: extracting payment period features from the insurance product data to be evaluated, thereby obtaining product payment period data;

In this embodiment, when extracting payment period features from the insurance product data to be evaluated, first extract payment period related data from the policy information. For example, fields such as the payment period and the first payment date in the policy can be parsed to establish product payment period data. By analyzing and sorting these data, the payment period features of the product are formed, including the range of payment periods, common payment periods, etc. For example, the payment period may be monthly, quarterly, or annual.

Step S42: Calculate the product default interaction frequency according to the product payment period data, thereby obtaining the default interaction frequency data of the product to be tested;

In this embodiment, the default interaction frequency of the product is calculated based on the product payment period data. The default interaction frequency of the product is calculated by analyzing the specific values of the payment period. For example, if the payment period is monthly, the default interaction frequency can be set to once a month. This can provide basic data for subsequent interaction experience evaluation.

Step S43: extracting product payment period features and interaction frequency features from historical insurance product data, thereby obtaining historical product payment period data and historical product actual interaction frequency data;

In this embodiment, the payment period feature extraction and interaction frequency feature extraction are performed on the historical data of insurance products. The historical payment period data and actual interaction frequency data of the product are extracted through the policy information in the historical data. This includes the payment period of the historical policy, the actual payment date and other information. Through these data, the payment period feature and actual interaction frequency feature of the historical product are formed.

Step S44: Conduct customer interaction experience evaluation on historical product payment period data, historical product actual interaction frequency data, and the default interaction frequency data of the product to be tested, thereby obtaining interaction experience evaluation data.

In this embodiment, the customer interaction experience is evaluated by combining the historical product payment period data, the historical product actual interaction frequency data, and the default interaction frequency data of the product to be tested. By comparing the default interaction frequency of the product to be tested with the actual interaction frequency of the historical product, it is evaluated whether the product interaction experience meets customer expectations. For example, if the default interaction frequency of the product to be tested is too high, and the customer prefers a lower actual interaction frequency, it may lead to a poor customer experience. The evaluation results can provide guidance for product improvement and form interaction experience evaluation data.

By extracting the payment period characteristics of the product, the present invention can understand the payment cycle and method of the product, and then provide basic data for subsequent interaction frequency calculation and customer interaction experience evaluation. It is helpful to evaluate the flexibility and customizability of the product, understand the customer's preferences in payment, and thus provide a reference for product customization and promotion. Calculating the default interaction frequency of the product to be tested helps to understand the degree of interaction between customers and products, and then provide a reference for product design and sales strategy. The frequency of use and customer participation of the product can be evaluated based on the interaction frequency data, providing guidance for product improvement and market positioning. By extracting the payment period characteristics and actual interaction frequency data of historical products, the characteristics of the product to be tested can be compared to understand the advantages and disadvantages of the product and its market competitiveness. It is helpful to discover the performance of historical products in terms of payment cycle and interaction frequency, and provide reference opinions for product improvement and optimization. Through customer interaction experience evaluation, customers' attitudes and preferences for product payment period and interaction frequency can be understood, thereby providing guidance for product improvement and market promotion. Customer satisfaction and experience feelings during actual use of the product can be evaluated, and improvement suggestions can be provided for product design and service improvement.

Optionally, step S44 is specifically:

Step S441: Calculate the default interaction frequency according to the historical product payment period data, thereby obtaining the historical product default interaction frequency data;

In this embodiment, the default interaction frequency of the historical product can be calculated through the payment period data of the historical product. For example, if the payment period of the historical product is once a month, then its default interaction frequency is once a month. This can be obtained by analyzing the frequency of the payment date in the historical data, thereby forming the default interaction frequency data of the historical product.

Step S442: performing a difference calculation on the historical product default interaction frequency data and the historical product actual interaction frequency data, thereby obtaining the historical product interaction difference data;

In this embodiment, the default interaction frequency data of the historical product is compared with the actual interaction frequency data to obtain the difference between them. This can be obtained by subtracting the default interaction frequency of the historical product from the actual interaction frequency. In this way, the interaction difference data of the historical product can be obtained to reflect the interaction frequency deviation of the historical product.

Step S443: extracting difference product data from historical insurance product data according to historical product interaction difference data, thereby obtaining interaction difference product data;

In this embodiment, the historical data of insurance products are processed using the interaction difference data of historical products to extract interaction difference product data. This can be achieved by applying a similar difference calculation method to apply the interaction difference of historical products to the historical data of other insurance products to obtain the corresponding interaction difference product data.

Step S444: performing similarity calculation on the interactive difference product data and the insurance product data to be evaluated, thereby obtaining product similarity data, and extracting high-similarity product data from the interactive difference product data according to the product similarity data, thereby obtaining high-similarity product data;

In this embodiment, by calculating the similarity between the interactive difference product data and the insurance product data to be evaluated, product similarity data can be obtained. This can be done by using various similarity calculation methods, such as cosine similarity or Euclidean distance. Then, based on the similarity data, products with high similarity are selected as references to obtain high similarity product data.

Step S445: Predicting the actual interaction frequency of the insurance product data to be evaluated based on the high-similarity product data, thereby obtaining actual interaction frequency prediction data of the product to be evaluated;

In this embodiment, based on the high-similarity product data, the actual interaction frequency of the insurance product data to be evaluated is predicted. This can be done by applying the actual interaction frequency of the high-similarity product to the product to be evaluated, thereby obtaining the actual interaction frequency prediction data of the product to be tested.

Step S446: Calculate the default interaction ratio based on the default interaction frequency data of the product to be tested and the actual interaction frequency prediction data of the product to be tested, so as to obtain interaction experience evaluation data.

In this embodiment, the default interaction ratio is calculated based on the default interaction frequency data of the product to be tested and the actual interaction frequency prediction data. This can be done by comparing the difference between the actual interaction frequency prediction value of the product to be tested and the default interaction frequency value, and then comparing the difference with the default interaction frequency to obtain the interaction experience evaluation data.

The present invention calculates the default interaction frequency through the payment period information in the historical data, understands the average customer interaction frequency of past products, and provides benchmark data for comparison and analysis. It is helpful to discover the general trend of products in past customer interactions and provide a reference basis for product design and market promotion. By calculating the difference between the default interaction frequency and the actual interaction frequency, the changes and trends of historical products in customer interaction can be identified, which helps to understand the actual use mode of customers. Interaction difference data for subsequent analysis is provided, which helps to discover the actual needs and expectations of customers for products. By extracting the interaction difference data of historical products, a set of product data reflecting the changes in customer interaction can be obtained, which helps to deeply understand the evolution process of products and the dynamics of customer interaction. Through similarity calculation, the similarity between historical products and products to be evaluated can be identified, providing a reference for product positioning, improvement and market promotion. Extracting high-similarity product data helps to find historical products with similar customer interaction patterns to the products to be evaluated, providing a basis for subsequent prediction and evaluation. Using the historical interaction data of high-similarity products, the actual customer interaction frequency of the products to be tested can be predicted, providing a basis for the expected interaction after the product is launched. It provides possible customer interaction patterns for the products to be tested in the future, which helps to formulate product design and market strategies. By calculating the ratio of default interaction frequency to actual interaction frequency, you can evaluate the customer interaction experience of the product and find out whether customers interact with the product as frequently as expected. This helps identify the product's advantages and improvement points in customer experience and provides a reference for product promotion and service improvement.

Optionally, the present specification further provides an insurance product evaluation system, which is used to execute the insurance product evaluation method as described above, and the insurance product evaluation system includes:

The risk assessment model building module is used to obtain historical data of insurance products and conduct claims risk assessment on the historical data of insurance products, thereby obtaining historical claims risk data; and building a claims risk assessment model based on the historical claims risk data;

The claim risk assessment module is used to obtain the insurance product data to be assessed, and to conduct claim risk assessment on the insurance product data to be assessed through the claim risk assessment model, thereby obtaining claim risk assessment data;

The coverage evaluation module is used to extract customer features from historical insurance product data to obtain historical customer data, and to classify historical customer data by customer loyalty to obtain customer loyalty data; and to conduct coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data to obtain coverage evaluation data;

The interactive experience evaluation module is used to calculate the product default interaction frequency of the insurance product data to be evaluated, thereby obtaining the default interaction frequency data of the product to be tested, and to evaluate the customer interactive experience of the default interaction frequency data of the product to be tested and the historical data of the insurance product, thereby obtaining the interactive experience evaluation data;

The comprehensive product evaluation module is used to conduct a comprehensive evaluation of insurance products based on claims risk evaluation data, coverage evaluation data, and interactive experience evaluation data, so as to obtain an insurance product evaluation report.

The insurance product evaluation system of the present invention can implement any insurance product evaluation method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the insurance product evaluation method. The internal modules of the system cooperate with each other, thereby improving the accuracy of the insurance product evaluation results.

Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims (10)

1. A method for evaluating an insurance product, comprising the following steps: Step S1: Obtain historical data of insurance products, and conduct claims risk assessment on the historical data of insurance products, thereby obtaining historical claims risk data; and construct a claims risk assessment model based on the historical claims risk data; Step S2: Obtain the insurance product data to be evaluated, and perform a claims risk evaluation on the insurance product data to be evaluated using a claims risk evaluation model, thereby obtaining claims risk evaluation data; Step S3: extracting customer features from historical data of insurance products to obtain historical customer data, and classifying historical customer data by customer loyalty to obtain customer loyalty data; conducting coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data to obtain coverage evaluation data; Step S4: Calculate the product default interaction frequency for the insurance product data to be evaluated, thereby obtaining the default interaction frequency data for the product to be tested, and perform customer interaction experience evaluation on the default interaction frequency data for the product to be tested and the insurance product historical data, thereby obtaining interaction experience evaluation data; Step S5: Perform a comprehensive evaluation of the insurance product based on the claims risk evaluation data, coverage evaluation data, and interactive experience evaluation data to obtain an insurance product evaluation report. 2. The insurance product evaluation method according to claim 1, characterized in that step S1 specifically comprises: Step S11: Obtaining historical data of insurance products; Step S12: extracting the term feature of the historical data of insurance products, thereby obtaining historical long-term insurance product data and historical short-term insurance product data; Step S13: conducting a long-term product claim risk assessment on historical long-term insurance product data, thereby obtaining long-term product claim risk data; Step S14: conducting a short-term product claim risk assessment on historical short-term insurance product data, thereby obtaining short-term product claim risk data; Step S15: merging the long-term product claim risk data and the short-term product claim risk data to obtain historical claim risk data; Step S16: Construct a claims risk assessment model based on historical claims risk data. 3. The insurance product evaluation method according to claim 2, characterized in that step S13 specifically comprises: Step S131: extracting features from historical long-term insurance product data, thereby obtaining historical long-term claims data and historical long-term premium payment data; Step S132: Calculate the claim probability of the same product based on the historical long-term claim data to obtain product claim probability data, and perform statistical analysis on the product claim probability data to obtain the product claim risk threshold; Step S133: classifying the historical long-term claims data according to the product claims risk threshold, thereby obtaining high claims risk long-term product data and low claims risk long-term product data; Step S134: Counting the number of missed payments by customers based on historical long-term premium payment data, thereby obtaining missed payment data of high-frequency customers and missed payment data of low-frequency customers; Step S135: classifying the historical long-term insurance product data by customer risk according to the high-frequency customer missed payment data and the low-frequency customer missed payment data, thereby obtaining high-customer risk long-term product data and low-customer risk long-term product data; Step S136: performing an intersection operation on the long-term product data with high claim risk and the long-term product data with high customer risk, thereby obtaining the long-term product claim high risk data; performing an intersection operation on the long-term product data with low claim risk and the long-term product data with low customer risk, thereby obtaining the long-term product claim low risk data; Step S137: Merge the long-term product claim low-risk data and the long-term product claim high-risk data to obtain the long-term product claim risk data. 4. The insurance product evaluation method according to claim 3, characterized in that step S14 specifically comprises: Step S141: extracting features from historical short-term insurance product data to obtain historical short-term claims data and customer age data; Step S142: performing claim frequency statistics on historical short-term claim data, thereby obtaining high-frequency short-term claim data and low-frequency short-term claim data; Step S143: performing classification calculation on the customer age data, thereby obtaining the age data of older customers and the age data of younger customers; Step S144: extracting age-restricted insurance products from historical short-term insurance product data based on the age data of senior customers and the age data of junior customers, thereby obtaining short-term age-restricted insurance product data; performing intersection operations on low-frequency short-term claims data and short-term age-restricted insurance product data, thereby obtaining short-term product claims low-risk data; Step S145: performing customer association on the high-frequency short-term claims data and the age data of senior customers, thereby obtaining first customer association data, and extracting high-risk senior customer product data from the historical short-term insurance product data based on the first customer association data, thereby obtaining high-risk data on senior short-term product claims; Step S146: performing customer association on the high-frequency short-term claims data and the age data of young customers, thereby obtaining second customer association data, and extracting high-risk young customer product data from the historical short-term insurance product data based on the second customer association data, thereby obtaining high-risk data on claims of young short-term products; Step S147: Merge the low-risk data on short-term product claims, the high-risk data on senior short-term product claims, and the high-risk data on junior short-term product claims to obtain short-term product claim risk data. 5. The insurance product evaluation method according to claim 1, characterized in that step S3 specifically comprises: Step S31: extracting customer features from historical insurance product data to obtain historical customer data; Step S32: extracting the insured product data from the historical customer data, thereby obtaining the customer's insured product data; Step S33: performing insurance period statistics on the customer insurance product data, thereby obtaining long-term customer insurance product data and short-term customer insurance product data, and calculating the insurance product ratio for the long-term customer insurance product data and short-term customer insurance product data, thereby obtaining customer insurance product ratio data; Step S34: Count the number of insured products based on the customer insurance data, thereby obtaining the number of insured products for the customer; Step S35: dividing the customer's insured product ratio data and the number of insured products by customer into customer loyalty categories, thereby obtaining customer loyalty data; Step S36: Conduct coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data, thereby obtaining coverage evaluation data. 6. The insurance product evaluation method according to claim 5, characterized in that the customer loyalty classification in step S35 is specifically: Conduct product ratio statistical analysis on the customer insurance product ratio data, so as to obtain customer data with a high proportion of short-term products, customer data with a high proportion of long-term products, customer data with monotonous products, and customer data with the same proportion; Conduct product quantity statistics analysis on the number of insured products purchased by customers, so as to obtain data on customers with high-value insured products and customers with low-value insured products; Perform customer intersection calculation based on the data of customers with a high proportion of long-term products and the data of customers with a high number of insurance premiums, so as to obtain the data of the first most loyal customers; Perform customer intersection calculation based on the data of customers with a high proportion of short-term products and the data of customers with a high number of insurance premiums, thereby obtaining the data of the second most loyal customers; Perform customer intersection calculation based on the data of customers with the same proportion and the data of customers with high insurance amount, so as to obtain the data of the third most loyal customers; Perform customer intersection calculation based on the monotonous product customer data and the high-value insurance customer data to obtain the fourth most loyal customer data; Perform customer intersection operations based on the monotonous product customer data and the low-amount insured customer data to obtain low-loyalty customer data; The first high-loyalty customer data, the second high-loyalty customer data, the third high-loyalty customer data and the fourth high-loyalty customer data are merged to obtain high-loyalty customer data, and the high-loyalty customer data and the low-loyalty customer data are merged to obtain customer loyalty data. 7. The insurance product evaluation method according to claim 5, characterized in that step S36 specifically comprises: Step S361: extracting features from the insurance product data to be evaluated, thereby obtaining the term data and premium data of the product to be evaluated; Step S362: extracting the customer's historical insurance product data from the insurance product history data according to the customer loyalty data, thereby obtaining the customer's historical insurance product data; Step S363: Calculate the product term proportion of the customer's historical insurance product data, thereby obtaining the customer's historical product term proportion data, and calculate the term coverage of the customer's historical product term proportion data according to the product term data to be evaluated, thereby obtaining the customer's term coverage data; Step S364: Calculate the product premium ratio of the customer's historical insurance product data, thereby obtaining the customer's historical product premium ratio data, and calculate the average premium of the customer's historical insurance product data, thereby obtaining the customer's average premium data; Step S365: Calculate the premium dispersion of the product to be evaluated and the average premium data of the customer, thereby obtaining premium dispersion data; calculate the premium coverage of the customer's historical product premium ratio data based on the premium dispersion data, thereby obtaining the customer's premium coverage data; Step S366: Evaluate the product coverage of the customer's historical insurance product data based on the customer's premium coverage data and the customer's term coverage data, thereby obtaining coverage evaluation data. 8. The insurance product evaluation method according to claim 1, characterized in that step S4 specifically comprises: Step S41: extracting payment period features from the insurance product data to be evaluated, thereby obtaining product payment period data; Step S42: Calculate the product default interaction frequency according to the product payment period data, thereby obtaining the default interaction frequency data of the product to be tested; Step S43: extracting product payment period features and interaction frequency features from historical insurance product data, thereby obtaining historical product payment period data and historical product actual interaction frequency data; Step S44: Conduct customer interaction experience evaluation on historical product payment period data, historical product actual interaction frequency data, and the default interaction frequency data of the product to be tested, thereby obtaining interaction experience evaluation data. 9. The insurance product evaluation method according to claim 8, characterized in that step S44 specifically comprises: Step S441: Calculate the default interaction frequency according to the historical product payment period data, thereby obtaining the historical product default interaction frequency data; Step S442: performing a difference calculation on the historical product default interaction frequency data and the historical product actual interaction frequency data, thereby obtaining the historical product interaction difference data; Step S443: extracting difference product data from historical insurance product data according to historical product interaction difference data, thereby obtaining interaction difference product data; Step S444: performing similarity calculation on the interactive difference product data and the insurance product data to be evaluated, thereby obtaining product similarity data, and extracting high-similarity product data from the interactive difference product data according to the product similarity data, thereby obtaining high-similarity product data; Step S445: Predicting the actual interaction frequency of the insurance product data to be evaluated based on the high-similarity product data, thereby obtaining actual interaction frequency prediction data of the product to be evaluated; Step S446: Calculate the default interaction ratio based on the default interaction frequency data of the product to be tested and the actual interaction frequency prediction data of the product to be tested, so as to obtain interaction experience evaluation data. 10. An insurance product evaluation system, characterized in that it is used to execute the insurance product evaluation method according to claim 1, and the insurance product evaluation system comprises: The risk assessment model building module is used to obtain historical data of insurance products and conduct claims risk assessment on the historical data of insurance products, thereby obtaining historical claims risk data; and building a claims risk assessment model based on the historical claims risk data; The claim risk assessment module is used to obtain the insurance product data to be assessed, and to conduct claim risk assessment on the insurance product data to be assessed through the claim risk assessment model, thereby obtaining claim risk assessment data; The coverage evaluation module is used to extract customer features from historical insurance product data to obtain historical customer data, and to classify historical customer data by customer loyalty to obtain customer loyalty data; and to conduct coverage evaluation on the insurance product data to be evaluated based on the customer loyalty data to obtain coverage evaluation data; The interactive experience evaluation module is used to calculate the product default interaction frequency of the insurance product data to be evaluated, thereby obtaining the default interaction frequency data of the product to be tested, and to evaluate the customer interactive experience of the default interaction frequency data of the product to be tested and the historical data of the insurance product, thereby obtaining the interactive experience evaluation data; The comprehensive product evaluation module is used to conduct a comprehensive evaluation of insurance products based on claims risk evaluation data, coverage evaluation data, and interactive experience evaluation data, so as to obtain an insurance product evaluation report.