Insurance customer-oriented insurance amount information analysis system

Abstract

The invention relates to an insurance customer-oriented insurance amount information analysis method and system, and the method comprises the steps: collecting customer information through a user interface, and carrying out the scanning through an OCR technology, and obtaining to-be-processed customer data; performing text extraction to obtain entities, relationships and keyword information, and performing grammar and semantic analysis; identifying insurance policy entities and relationships by using a machine learning technology, and organizing the insurance policy entities and relationships into structured data; according to the insurance information category, establishing a plurality of customer information models based on the structured data; establishing customer feature tags for each customer information model by using a machine learning algorithm, performing customer classification, and predicting behaviors and potential risks of customers; and quantizing the insurance amount gap by adopting a gap deduction formula of a corresponding insurance type, providing a corresponding insurance guarantee suggestion, and presenting the insurance guarantee suggestion in a visual report form. Compared with the prior art, the insurance amount analysis is more intelligent and accurate, and more personalized and specialized insurance guarantee can be provided for customers.

Description

Technical Field

[0001] The present invention relates to the technical field of insurance data processing, and in particular to an insurance amount information analysis system for insurance customers. Background Art

[0002] In today's rapidly developing insurance industry, customer management with thousands of different faces has become the core strategy of insurance companies. In this context, customer demand analysis is particularly critical, and protection gap analysis has become an important means to achieve customer management with thousands of different faces. With the rapid development of financial technology, insurance products are constantly innovating, and customer needs are also showing a trend of becoming more diversified and personalized.

[0003] When salesmen conduct customer analysis, protection gap analysis plays a vital role. Salesmen need to have a deep understanding of customers' insurance needs to ensure that they can provide them with the best protection plan. Protection gap analysis provides salesmen with more accurate information on customer needs by conducting an in-depth analysis of the gap between customers' existing insurance coverage and actual needs.

[0004] This kind of customer demand analysis based on individual needs can help salesmen understand customers' insurance needs more comprehensively, provide more targeted consulting services, and also help to build closer customer relationships. In the context of increasingly fierce market competition, customer management based on individual needs has become a key link in the intelligent transformation of the insurance industry. Therefore, it is necessary to ensure the insurance services that are lacking. Through customer management based on individual needs, insurance companies can better adapt to the diversified needs of the market and improve customer satisfaction and loyalty.

[0005] However, there are three problems with the insurance coverage gap analysis currently conducted by the insurance industry: the insurance coverage analysis model is single, the breadth and depth of customer information is insufficient, and agents lack professional knowledge.

[0006] First, the current insurance coverage analysis mainly relies on basic information such as age, gender, income, etc., which is directly mapped to insurance products, without truly in-depth gap analysis based on the unique situation of customers. It lacks comprehensive consideration of customers' personalized needs, and needs to shift to a more in-depth analysis, focusing on customers' unique needs and special backgrounds, in order to develop more realistic insurance coverage plans, rather than simply applying general templates.

[0007] Secondly, the current information collection is still relatively limited, staying at basic information and family information, and failing to complete the comprehensive modeling of customers. In order to conduct a more accurate analysis of the gap in family insurance coverage, it is necessary to strengthen the in-depth collection of customer information, including occupation, family relationships, health, financial status, existing insurance coverage, etc. Only comprehensive modeling can more accurately reveal the gap in customers' insurance needs.

[0008] In addition, when conducting insurance coverage analysis, salespeople need a wide range of professional knowledge, including financial management, insurance products, statistical analysis, etc. Existing training and learning opportunities fail to meet this broad demand, making it difficult for salespeople to master the necessary professional knowledge. However, the current education level of the insurance agent team is relatively low, and even if there is an opportunity to learn, it is difficult for them to understand these professional capabilities. Summary of the invention

[0009] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide an insurance coverage information analysis system for insurance customers, which can efficiently collect, structure and analyze customer information and provide customers with more accurate and personalized insurance recommendations.

[0010] The purpose of the present invention can be achieved by the following technical solutions:

[0011] A method for analyzing insurance sum insured information for insurance customers, comprising the following steps:

[0012] Customer information collection steps: collect customer information through the user interface, the customer information includes basic information, financial information, health information, existing insurance information, comprehensive risk assessment information, and insurance planning information; use OCR technology to scan the electronic insurance policies and paper insurance policies in the customer information, and combine them with the existing insurance information to obtain the customer data to be processed; perform text extraction on the customer data to be processed to obtain entity, relationship and keyword information, and perform grammatical and semantic analysis on the extracted text; use machine learning technology to identify the policy entities and relationships of the extraction and analysis results, and organize them into structured data for insurance;

[0013] Customer information modeling steps: According to the insurance information category, establish multiple customer information models based on the acquired structured data; use machine learning algorithms for each customer information model to establish customer feature labels and classify customers; and predict customer behavior and potential risks respectively;

[0014] Steps for analyzing customer protection gaps: Based on the data from each customer information model, use the pre-built gap deduction formula for the corresponding insurance type to quantify the insured amount gap, provide corresponding insurance protection recommendations, and present them in the form of a visual report.

[0015] Furthermore, the text extraction process specifically includes:

[0016] Entity identification sub-step: Use NLP technology to perform entity identification on the customer data to be processed, where the entities include insurance type name, liability name, date and amount;

[0017] Relationship extraction step: Based on the identified entities, determine the relationship between entities;

[0018] Keyword extraction step: extract the keywords of the insurance contract from the customer data to be processed.

[0019] Furthermore, the process of grammatical and semantic analysis specifically includes:

[0020] Syntactic analysis substep: determining the grammatical relationships between words in the text;

[0021] Semantic role labeling sub-step: determine the semantic roles of each part of the sentence in the text and label them;

[0022] Word embedding sub-step: Map the words in the text into a high-dimensional vector space to capture the semantic relationship between words.

[0023] Furthermore, the identification process of the insurance policy entities and relationships is specifically as follows:

[0024] A named entity recognition model and a relationship extraction model are constructed respectively through machine learning technology. The named entity recognition model is trained based on labeled insurance field data and is used to identify policy entities based on the extraction and analysis results; the relationship extraction model is used to capture the relationship between different policy entities based on the identified policy entities.

[0025] Furthermore, the organization process of the structured data is specifically as follows:

[0026] A data model is defined according to entities and the relationships between entities, and the identified policy entities and relationships are constructed in the form of the data model to obtain the structured data.

[0027] Furthermore, the customer information collection step also includes storing the obtained structured insurance data in a corresponding database for data retrieval and query.

[0028] Furthermore, the established customer information model includes:

[0029] Basic information model, which includes personal information, including age, gender, marital status and education level; family information, including the number of family members, family structure and place of residence; occupational information, including occupation type, nature of work and occupational risks;

[0030] Financial information model, which includes income level, including annual income of individuals and families; debt situation, including loans, debts and credit status; financial goals, including the client's investment plan, financial management needs and financial goals;

[0031] Health information model, which includes health status, including the client's physical indicators and health assessment results; medical history, including past illnesses, surgeries and treatment information; lifestyle habits, including diet, exercise, and smoking habits;

[0032] An existing insurance information model, which includes existing insurance policies, including the types and quantities of insurance products purchased; the validity period of insurance policies, including the validity period of existing insurance policies; and the scope of protection, including the scope of risks covered by existing insurance policies;

[0033] A comprehensive risk assessment model, which includes comprehensive risk tolerance, including the customer's tolerance for financial and health risks; protection demand priorities, including the customer's priority ranking results for insurance protection;

[0034] The insurance planning model includes education planning, including the financial needs of children's education; retirement planning, including the client's pension plan, retirement goals and expected income; major event planning, including financial preparation plans for accidents and major illnesses.

[0035] Furthermore, the process of establishing the customer feature labels includes: using machine learning algorithms based on indicators in various customer information models to obtain customer classification results;

[0036] The prediction process of the customer's behavior and potential risks is specifically as follows:

[0037] Based on the information in each customer information model, the customer behavior and potential risks are predicted through the pre-trained prediction model;

[0038] The customer information modeling step also includes: updating the information in each customer information model in real time, and monitoring the performance of the machine learning model in the process of establishing customer feature labels and predicting customer behavior and potential risks in real time.

[0039] Furthermore, the gap deduction formula includes: a medical insurance coverage gap analysis formula, a critical illness insurance coverage gap analysis formula, a term life insurance coverage gap analysis formula, an accident insurance coverage gap analysis formula, a pension coverage gap analysis formula, and an education fund coverage gap analysis formula;

[0040] The calculation expression of the medical insurance coverage gap analysis formula is:

[0041] Gap amount = (estimated medical expenditure coefficient * age coefficient * health status coefficient) - existing medical insurance coverage. The calculation expression of the critical illness insurance coverage gap analysis formula is:

[0042] Gap amount = (estimated treatment cost factor * health condition factor * disease risk factor) - existing critical illness insurance coverage

[0043] The calculation expression of the term life insurance coverage gap analysis formula is:

[0044] Gap amount = (family financial demand coefficient * age coefficient * financial target coefficient) - existing term life insurance coverage. The calculation expression of the accident insurance coverage gap analysis formula is:

[0045] Gap amount = (accidental injury treatment cost coefficient * living habits coefficient * work environment coefficient) - existing accident insurance coverage

[0046] The calculation expression of the pension insurance gap analysis formula is:

[0047] Gap amount = (expected pension expenditure coefficient * retirement age coefficient * expected living standard coefficient) - existing pension plan

[0048] The calculation expression of the education fund insured amount gap analysis formula is:

[0049] Gap amount = (children's education expense coefficient * education stage coefficient * tuition level coefficient) - existing education fund plan.

[0050] The present invention also provides an insurance amount information analysis system based on the insurance amount information analysis method for insurance customers as described above, comprising:

[0051] The customer information collection module is used to scan the customer's electronic and paper insurance policies using OCR technology, and combine the existing insurance information to obtain the customer data to be processed; extract the text of the customer data to be processed to obtain entity, relationship and keyword information, and perform grammatical and semantic analysis on the extracted text; use machine learning technology to identify the policy entities and relationships of the extraction and analysis results, and organize them into structured insurance data;

[0052] The customer information modeling module is used to establish multiple customer information models based on the acquired structured data according to the insurance information category; use machine learning algorithms to establish customer feature labels for each customer information model to classify customers; and predict customer behaviors and potential risks respectively;

[0053] The customer protection gap analysis module is used to quantify the insured amount gap based on the data of each customer information model, adopt the pre-built gap deduction formula of the corresponding insurance type, provide corresponding insurance protection suggestions, and present them in the form of a visual report.

[0054] Compared with the prior art, the present invention has the following advantages:

[0055] (1) In the process of analyzing the insured amount of insurance customers, the present invention first integrates OCR technology through the customer information collection stage to obtain personal information, financial status, health information and other data in a secure and encrypted manner. Subsequently, through the customer information modeling step, a comprehensive customer information model is constructed from multiple perspectives, including basic information, financial information, health information, etc., and machine learning technology is used to ensure a comprehensive grasp of customer characteristics. Finally, through customer protection gap analysis, analysis standards and deduction formulas for six major insurance types are set, the insured amount gap is quantitatively analyzed, a visual analysis report is provided, and an intelligent recommendation system is introduced to provide customers with personalized insurance plan suggestions. These three steps together constitute a systematic and complete insured amount analysis process, with the goal of meeting the individual needs of customers.

[0056] (2) The present invention breaks through the original information acquisition model and abandons the narrow form of form filling. Through relevant processing technology, it can realize data collection of multiple business processes such as questionnaires, assessments and physical examinations, and realize multi-faceted and multi-level information collection of customers, truly realizing a comprehensive understanding of customers' goals and needs. Especially in the OCR collection of insurance policies, through the application of cutting-edge technology, key information is automatically extracted, and the insurance policy data is structured, providing a more accurate and comprehensive data basis for subsequent insurance coverage analysis.

[0057] In terms of customer information modeling, we have gotten rid of the simple reliance of traditional models on basic, financial and health information, and built a more comprehensive and three-dimensional information model with customers as the center, including multi-level data such as existing insurance information, comprehensive risk assessment information and insurance planning information. This comprehensive modeling method provides a richer and more specific basis for subsequent analysis.

[0058] Finally, in the insurance coverage analysis stage, we no longer apply simple scorecard mapping, but adopt scientific actuarial formulas. Combined with the detailed data of the customer information model, we can provide each customer with a personalized insurance analysis report. This personalized and visual presentation method provides customers with a more intuitive and in-depth understanding, allowing them to better understand the necessity and rationality of insurance planning.

[0059] (3) In summary, the present invention has injected new vitality into the insurance industry through innovations in multiple links such as information collection, modeling and analysis in the entire insurance coverage analysis process, providing customers with more comprehensive and personalized insurance planning services and realizing true insurance customization. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of the structure of a system for analyzing insurance coverage information for insurance customers provided in an embodiment of the present invention;

[0061] Figure 2The present invention is a flowchart of a method for analyzing insurance coverage information for insurance customers provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0063] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0065] Example 1

[0066] In response to the problems existing in the prior art, this embodiment provides an insurance coverage information analysis system for insurance customers, so as to fundamentally solve the problems of general templated insurance coverage analysis, limited customer information modeling, and difficulty in cultivating agent professional capabilities currently faced by the insurance industry. This system should have advanced personalized algorithms to more accurately meet the needs of different customers and get rid of traditional general routines. At the same time, the system should realize comprehensive customer information collection and modeling, deeply mine multi-dimensional information, and provide more in-depth data support for insurance coverage analysis. More importantly, the system design should simplify the understanding of professional knowledge and reduce the difficulty of operation for agents, so as to truly promote the realization of personalized services in the industry. Such an innovative system will bring revolutionary changes to the insurance industry, making insurance coverage analysis more intelligent and accurate, and providing customers with more personalized and professional insurance protection.

[0067] like Figure 1As shown in the figure, the customer information collection module, customer information modeling module and customer protection gap analysis module are specifically designed. Through the design of the three major sections, the system can efficiently collect, structure and analyze customer information, and provide customers with more accurate and personalized insurance advice. Through the OCR technology, intelligent scanning of insurance policy information is realized, and data matching is carried out in combination with the industry responsibility database, so that unstructured existing insurance information can be effectively utilized. The significance of the system is to improve the professionalism of insurance sales, while meeting the actual needs of customers for insurance coverage and increasing customers' trust in insurance products. The implementation of this system will accelerate the digital transformation of the insurance industry, improve operational efficiency, and provide customers and insurance practitioners with a better experience.

[0068] The processing procedures of the above modules are as follows:

[0069] 1. Customer Information Collection Module

[0070] First, the present invention establishes a customer information collection process system, designs a user-friendly interface, creates a good process experience, and guides customers to provide six categories of information: basic information, financial information, health information, existing insurance information, comprehensive risk assessment information, and insurance planning information.

[0071] Secondly, OCR technology is integrated to support electronic and paper policy scanning for existing insurance information of customers. Natural language processing (NLP) algorithms are used to process electronic or paper policies. The unstructured information is structured through the following steps. The specific process is as follows:

[0072] 1. Text extraction: Use NLP technology to extract text from the customer data to be processed to obtain entity, relationship and keyword information;

[0073] 1.1. Entity Recognition (NER): Through model training, NLP can accurately identify named entities in text, such as insurance type names (such as medical insurance, critical illness insurance), liability names (general medical care, critical illness medical care), dates (effective date, expiration date), amounts (insured amount, premium), etc.;

[0074] 1.2. Relationship extraction: determine the relationship between the identified entities, such as the association between the insurance type name and the insured amount;

[0075] 1.3. Keyword extraction: Extracting keywords from insurance contracts helps to better understand the content of the contract.

[0076] 2. Syntax and semantic analysis: Perform grammatical and semantic analysis on the extracted text to gain a deeper understanding of the structure and meaning of the text;

[0077] 2.1. Syntactic analysis: determine the grammatical relationship between words in the text, such as subject, predicate, object, etc.;

[0078] 2.2. Semantic role labeling: Determine the semantic roles of various parts in a sentence to help understand actions and events;

[0079] 2.3. Word Embedding: Map the words in the policy text into a high-dimensional vector space to capture the semantic relationships between them.

[0080] 3. Model training and optimization: Use machine learning technology, especially deep learning models, to train and optimize the extracted information to realize the identification of insurance policy entities and relationships; specifically, you can build a named entity recognition model and a relationship extraction model respectively.

[0081] 3.1. Named Entity Recognition Model: Use labeled insurance data for training to ensure that the model accurately recognizes policy entities such as insurance type name, date, amount, etc.

[0082] 3.2. Relationship extraction model: Based on the extracted entities, the model is trained to capture the relationship between different entities, such as the connection between insurance type and insured amount.

[0083] 4. Data structuring: Organize the extracted information into structured data to facilitate subsequent analysis.

[0084] 4.1. Create data model: Define the data model, including entities and their relationships, in order to establish a clear policy information structure;

[0085] 4.2. Database storage: Store structured information in a specially designed database to facilitate retrieval and query in subsequent insurance coverage analysis.

[0086] Through these steps, the NLP algorithm is used to convert the unstructured information in electronic or paper policies into structured insurance data, providing a data basis for subsequent insurance coverage analysis.

[0087] 2. Customer Information Modeling Module

[0088] It is used to establish multiple customer information models based on the acquired structured data according to the insurance information category; use machine learning algorithms to establish customer feature labels for each customer information model and classify customers; and predict customer behaviors and potential risks respectively;

[0089] 1. Establish customer information model

[0090] 1.1 Basic information model:

[0091] Personal information: including age, gender, marital status, education level, etc.

[0092] Family information: Consider the number of family members, family structure, place of residence, etc.

[0093] Occupational information: collect detailed information on occupation type, nature of work, occupational risks, etc.

[0094] 1.2 Financial Information Model:

[0095] Income level: covers individual and family annual income, including the main source of income.

[0096] Debt status: Consider loans, debts, credit status, etc.

[0097] Financial goals: Get a detailed understanding of the client’s investment plans, financial management needs and financial goals.

[0098] 1.3. Health Information Model:

[0099] Health status: Collect information about the client’s overall health status, which may include physical indicators and health assessments.

[0100] Medical history: Learn about past illnesses, surgeries, and treatments to assess medical risks.

[0101] Lifestyle habits: Have a detailed understanding of the impact of lifestyle habits such as diet, exercise, and smoking.

[0102] 1.4. Existing insurance information model:

[0103] Existing policies: Collect the types and quantities of insurance products purchased.

[0104] Policy validity period: Understand the validity period of existing policies and whether they need to be updated or adjusted.

[0105] Coverage: Understand in detail the scope of risks covered by existing policies to ensure comprehensive protection.

[0106] 1.5. Comprehensive risk assessment model:

[0107] Comprehensive risk tolerance: Gain insight into your client’s tolerance for different risks, including financial and health risks.

[0108] Prioritization of protection needs: Discuss with customers the priority of various insurance protections and help develop personalized protection plans.

[0109] 1.6. Insurance planning model:

[0110] Education planning: Understand the financial needs of children’s education in order to develop relevant insurance plans.

[0111] Retirement planning: Get a detailed understanding of your client’s retirement plans, retirement goals, and expected income.

[0112] Major Event Planning: Discuss financial preparedness plans for major events such as accidents and major illnesses.

[0113] 2. Apply machine learning and data mining techniques

[0114] 2.1. Establishing customer feature labels: Based on the indicators in various customer information models, machine learning algorithms are used to obtain customer classification results; specifically:

[0115] Using machine learning algorithms, customer feature labels are established based on various indicators in the customer information model.

[0116] Use supervised learning models, such as classification algorithms, to categorize customers, such as risk preference, purchasing power, etc.

[0117] 2.2. Establish a prediction model: Based on the information in each customer information model, the customer's behavior and potential risks are predicted through the pre-trained prediction model; specifically:

[0118] Establish prediction models according to different model requirements, such as risk prediction, purchase behavior prediction, etc.

[0119] Use machine learning algorithms such as regression and classification to train models to predict customer behavior or potential risks.

[0120] 2.3. Real-time update and monitoring: Update the information in each customer information model in real time, and monitor the performance of the machine learning model in the process of establishing customer feature labels and predicting customer behavior and potential risks in real time; specifically:

[0121] Realize real-time updating of the model to promptly reflect changes in customer information.

[0122] Establish a monitoring mechanism to monitor the performance of the model in real time to ensure the effectiveness of the model in different environments.

[0123] 3. Customer Protection Gap Analysis

[0124] It is used to quantify the insured amount gap based on the data of each customer information model, adopt the pre-built gap deduction formula of the corresponding insurance type, provide corresponding insurance protection suggestions, and present them in the form of a visual report.

[0125] 1. Set the analysis standards and deduction formulas for the six major types of insurance:

[0126] 1.1. Medical insurance coverage gap analysis formula:

[0127] Gap amount = (estimated medical expenditure coefficient * age coefficient * health status coefficient) - current medical insurance coverage

[0128] Among them, each coefficient needs to be adjusted and calculated according to the specific data in the customer information model.

[0129] 1.2. Critical illness insurance coverage gap analysis formula:

[0130] Gap amount = (estimated treatment cost factor * health condition factor * disease risk factor) - existing critical illness insurance coverage

[0131] Each coefficient also needs to be adjusted and calculated based on the specific data in the customer information model.

[0132] 1.3、Term life insurance coverage gap analysis formula:

[0133] Gap amount = (family financial needs coefficient * age coefficient * financial goal coefficient) - existing term life insurance coverage

[0134] Each coefficient needs to be adjusted and calculated based on the specific data in the customer financial information model and the household information model.

[0135] 1.4. Accident insurance coverage gap analysis formula:

[0136] Gap amount = (accidental injury treatment cost coefficient * living habits coefficient * work environment coefficient) - existing accident insurance coverage

[0137] Each coefficient needs to be adjusted and calculated based on specific data such as the customer's health information model and living habits.

[0138] 1.5. Pension insurance gap analysis formula:

[0139] Gap amount = (expected pension expenditure coefficient * retirement age coefficient * expected living standard coefficient) - existing pension plan

[0140] Each coefficient needs to be adjusted and calculated based on the specific data in the customer financial information model and personal information model.

[0141] 1.6、Education fund insured amount gap analysis formula:

[0142] Gap amount = (children's education cost coefficient * education stage coefficient * tuition level coefficient) - existing education fund plan

[0143] Each coefficient needs to be adjusted and calculated based on the specific data in the customer's household information model and financial information model.

[0144] Such a detailed formula takes into account the impact of various factors and can more accurately analyze the coverage gap for each type of insurance.

[0145] 2. Conduct quantitative analysis of the insured amount gap:

[0146] 2.1. Apply the data in the customer information model to each deduction formula to achieve specific quantification of the insurance coverage gap.

[0147] 2.2. Use machine learning and data mining techniques to analyze large amounts of data to identify potential under-insurance or over-insurance situations.

[0148] 2.3. Provide customers with comprehensive insurance protection recommendations to ensure that the coverage of each type of insurance matches customer needs.

[0149] 3. Provide visual analysis reports:

[0150] Produce intuitive and easy-to-understand visual reports to present the insured gap of each type of insurance. Display through charts, graphs, etc. to help customers understand insurance needs and suggestions more clearly. The report contains detailed data explanations and suggestions to support customers' understanding and decision-making.

[0151] Example 2

[0152] like Figure 2 As shown, this embodiment provides a method for analyzing insurance coverage information for insurance customers, comprising the following steps:

[0153] Customer information collection step S1: collect customer information through the user interface, including basic information, financial information, health information, existing insurance information, comprehensive risk assessment information, and insurance planning information; use OCR technology to scan the electronic insurance policies and paper insurance policies in the customer information, and combine them with the existing insurance information to obtain the customer data to be processed; perform text extraction on the customer data to be processed to obtain entity, relationship and keyword information, and perform grammatical and semantic analysis on the extracted text; use machine learning technology to identify the policy entities and relationships of the extraction and analysis results, and organize them into structured data of insurance;

[0154] Customer information modeling step S2: According to the insurance information category, multiple customer information models are established based on the acquired structured data; machine learning algorithms are used for each customer information model to establish customer feature labels and classify customers; and customer behaviors and potential risks are predicted respectively;

[0155] Customer protection gap analysis step S3: Based on the data of each customer information model, the pre-built gap deduction formula of the corresponding insurance type is used to quantify the insurance coverage gap, provide corresponding insurance protection suggestions, and present them in the form of a visual report.

[0156] The above steps correspond to the specific processing procedures of the customer information collection module, customer information modeling module and customer protection gap analysis module for insurance customer's insurance coverage information analysis in Example 1, which will not be repeated here.

[0157] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A method for analyzing insurance sum insured information for insurance customers, characterized in that: The following steps are involved: Customer information collection steps: collect customer information through the user interface, the customer information includes basic information, financial information, health information, existing insurance information, comprehensive risk assessment information, and insurance planning information; use OCR technology to scan the electronic insurance policies and paper insurance policies in the customer information, and combine them with the existing insurance information to obtain the customer data to be processed; perform text extraction on the customer data to be processed to obtain entity, relationship and keyword information, and perform grammatical and semantic analysis on the extracted text; use machine learning technology to identify the policy entities and relationships of the extraction and analysis results, and organize them into structured data for insurance; Customer information modeling steps: According to the insurance information category, establish multiple customer information models based on the acquired structured data; use machine learning algorithms for each customer information model to establish customer feature labels, classify customers, and predict customer behaviors and potential risks; Steps for analyzing customer protection gaps: Based on the data from each customer information model, use the pre-built gap deduction formula for the corresponding insurance type to quantify the insured amount gap, provide corresponding insurance protection recommendations, and present them in the form of a visual report.

2. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The text extraction process specifically includes: Entity identification sub-step: Use NLP technology to perform entity identification on the customer data to be processed, where the entities include insurance type name, liability name, date and amount; Relationship extraction step: Based on the identified entities, determine the relationship between entities; Keyword extraction step: extract the keywords of the insurance contract from the customer data to be processed.

3. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The process of grammatical and semantic analysis specifically includes: Syntactic analysis substep: determining the grammatical relationships between words in the text; Semantic role labeling sub-step: determine the semantic roles of each part of the sentence in the text and label them; Word embedding sub-step: Map the words in the text into a high-dimensional vector space to capture the semantic relationship between words.

4. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The identification process of the insurance policy entities and relationships is specifically as follows: A named entity recognition model and a relationship extraction model are constructed respectively through machine learning technology. The named entity recognition model is trained based on labeled insurance field data and is used to identify policy entities based on the extraction and analysis results; the relationship extraction model is used to capture the relationship between different policy entities based on the identified policy entities.

5. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The organization process of the structured data is specifically as follows: A data model is defined according to entities and the relationships between entities, and the identified policy entities and relationships are constructed in the form of the data model to obtain the structured data.

6. A method for analyzing insurance customer's insurance coverage information according to claim 5, characterized in that: The customer information collection step also includes storing the obtained structured insurance data in a corresponding database for data retrieval and query.

7. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The customer information model established includes: Basic information model, which includes personal information, including age, gender, marital status and education level; family information, including the number of family members, family structure and place of residence; occupational information, including occupation type, nature of work and occupational risks; Financial information model, which includes income level, including annual income of individuals and families; debt situation, including loans, debts and credit status; financial goals, including the client's investment plan, financial management needs and financial goals; Health information model, which includes health status, including the client's physical indicators and health assessment results; medical history, including past illnesses, surgeries and treatment information; lifestyle habits, including diet, exercise, and smoking habits; An existing insurance information model, which includes existing insurance policies, including the types and quantities of insurance products purchased; the validity period of insurance policies, including the validity period of existing insurance policies; and the scope of protection, including the scope of risks covered by existing insurance policies; A comprehensive risk assessment model, which includes comprehensive risk tolerance, including the customer's tolerance for financial and health risks; protection demand priorities, including the customer's priority ranking results for insurance protection; The insurance planning model includes education planning, including the financial needs of children's education; retirement planning, including the client's pension plan, retirement goals and expected income; major event planning, including financial preparation plans for accidents and major illnesses.

8. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The process of establishing the customer feature label includes: based on the indicators in various customer information models, using machine learning algorithms to obtain customer classification results; The prediction process of the customer's behavior and potential risks is specifically as follows: Based on the information in each customer information model, the customer behavior and potential risks are predicted through the pre-trained prediction model; The customer information modeling step also includes: updating the information in each customer information model in real time, and monitoring the performance of the machine learning model in the process of establishing customer feature labels and predicting customer behavior and potential risks in real time.

9. A method for analyzing insurance customer's insurance coverage information according to claim 1, characterized in that: The gap deduction formulas include: medical insurance coverage gap analysis formula, critical illness insurance coverage gap analysis formula, term life insurance coverage gap analysis formula, accident insurance coverage gap analysis formula, pension coverage gap analysis formula and education fund coverage gap analysis formula; The calculation expression of the medical insurance coverage gap analysis formula is: Gap amount = (estimated medical expenditure coefficient * age coefficient * health status coefficient) - existing medical insurance coverage. The calculation expression of the critical illness insurance coverage gap analysis formula is: Gap amount = (estimated treatment cost factor * health condition factor * disease risk factor) - existing critical illness insurance coverage The calculation expression of the term life insurance coverage gap analysis formula is: Gap amount = (family financial demand coefficient * age coefficient * financial target coefficient) - existing term life insurance coverage. The calculation expression of the accident insurance coverage gap analysis formula is: Gap amount = (accidental injury treatment cost coefficient * living habits coefficient * work environment coefficient) - existing accident insurance coverage The calculation expression of the pension insurance gap analysis formula is: Gap amount = (expected pension expenditure coefficient * retirement age coefficient * expected living standard coefficient) - existing pension plan The calculation expression of the education fund insured amount gap analysis formula is: Gap amount = (children's education expense coefficient * education stage coefficient * tuition level coefficient) - existing education fund plan.

10. An insurance amount information analysis system based on the insurance amount information analysis method for insurance customers as claimed in any one of claims 1 to 9, characterized in that: include: The customer information collection module is used to scan the customer's electronic and paper insurance policies using OCR technology, and combine the existing insurance information to obtain the customer data to be processed; extract the text of the customer data to be processed to obtain entity, relationship and keyword information, and perform grammatical and semantic analysis on the extracted text; use machine learning technology to identify the policy entities and relationships of the extraction and analysis results, and organize them into structured insurance data; The customer information modeling module is used to establish multiple customer information models based on the acquired structured data according to the insurance information category; use machine learning algorithms to establish customer feature labels for each customer information model to classify customers; and predict customer behaviors and potential risks respectively; The customer protection gap analysis module is used to quantify the insured amount gap based on the data of each customer information model, adopt the pre-built gap deduction formula of the corresponding insurance type, provide corresponding insurance protection suggestions, and present them in the form of a visual report.

Images