Intelligent safety responsibility insurance buying information management method and system

Abstract

The invention discloses an intelligent safety responsibility insurance buying information management method and system, and particularly relates to the field of data processing and analysis, and the method comprises the steps: S1, multi-source data fusion and collection, S2, multi-source data processing and mining, S3, pricing model construction, S4, intelligent buying recommendation, and S5, automatic insurance policy management. According to the method, a mathematical model is constructed by comprehensively applying multiple data sources and an advanced data analysis method, text data related to an enterprise is deeply analyzed by applying a natural language processing technology, and meanwhile, multi-dimensional factors such as the enterprise scale, the industry risk coefficient, the equipment operation health degree and the safety management system perfection degree are combined; and along with continuous input of new data, the model can automatically learn and optimize, enterprise risk changes can be dynamically tracked in real time, and risk levels and probability evaluation which better conform to actual conditions are provided.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and analysis, and more specifically, to an intelligent method and system for managing occupational liability insurance information. Background Art

[0002] Application No. CN202211578668.5 discloses a method for determining the premium of construction liability insurance, which relates to the field of construction engineering safety technology, and in particular to a method for determining the premium of construction liability insurance. In view of the problems that it is difficult to conduct comprehensive evaluation of construction projects in current technology, the steps for determining the premium of liability insurance are cumbersome, the evaluation results are vague, and the premium cannot be adjusted in time, the following scheme is proposed, including the following steps: S1: Using a static model integrating LECD method and fuzzy mathematics to conduct risk assessment on construction projects; the purpose of the present invention is to conduct risk assessment on construction projects and provide a method for determining the premium of enterprise and project liability insurance. By adopting a static model based on the combination of improved LECD and fuzzy mathematics, multi-level weight allocation, further analyzing risk information, proposing a risk matrix for construction projects, combining the insurance premium matrix, providing the insurance premium of projects and enterprises, and conducting dynamic evaluation in real time, timely adjusting the premium, and improving the efficiency of premium determination.

[0003] Application No. CN202110153043.3 discloses a safety liability insurance underwriting system and method based on blockchain technology, which relates to a safety liability insurance underwriting system and method based on blockchain technology. Among them, the safety liability insurance underwriting system based on blockchain technology includes: a data perception layer, a blockchain network layer, a data monitoring system, a risk analysis system, an accident prevention service system, and a policy management system. In the present invention, the data perception layer screens and uploads the basic data of the insured enterprises to the chain, providing effective and reliable data support for other system modules; the blockchain network layer is used to effectively reduce various friction costs in the process of information transmission, and enhance the degree of trust between the participants; it meets the actual needs of the insured enterprises for providing accident prevention services, thereby improving the safety production level of the enterprises; and meets the needs of the regulatory authorities for providing risk analysis services for enterprises under their jurisdiction.

[0004] However, there are still some shortcomings in its actual use. For example, existing technologies mostly rely on limited information regularly reported by enterprises and some offline survey data, and the data types are mostly numerical data, which will lead to insufficient data and affect subsequent pricing analysis. The data regularly reported by enterprises are often limited to some basic enterprise profile information, such as company name, registered address, business scope, number of employees, etc. This information can only outline the general outline of the enterprise, and it is not enough for in-depth insight into the complex production safety operation status within the enterprise. Although offline survey data can supplement some information to a certain extent, due to factors such as high investigation costs, limited coverage and low investigation frequency, the amount of data obtained is still very limited, which will lead to insufficient collection of data and affect subsequent pricing analysis. At the same time, traditional risk assessment models are often based on numerical data and simple statistical analysis or empirical rules, and the assessment results are not accurate enough. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an intelligent method and system for managing occupational liability insurance information, and solve the problems raised in the above-mentioned background technology through the following scheme.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent method and system for the management of occupational liability insurance information, including: S1: multi-source data fusion collection, S2: multi-source data processing and mining, S3: pricing model construction, S4: intelligent insurance recommendation and S5: automated policy management; The S1 multi-source data fusion collection: using the Internet of Things technology, deploying sensors in the enterprise's production equipment and working environment, collecting equipment operation data, environmental parameters and employee operation behavior data in the production process in real time, connecting with the database of the safety supervision department, obtaining the enterprise's safety inspection records, administrative penalty information, industry safety standards and specifications data, integrating the enterprise's internal safety management system data, including safety management system documents, safety training records, hidden danger investigation records and rectification ledger records, and collecting safety production accident information in the industry through web crawler technology; The S2 multi-source data processing and mining: pre-processing the collected multi-source data, cleaning the collected text data, removing numbers, dates, advertisements, emoticons, hyperlinks and modal particles, extracting words from sentences according to the dictionary and classifying the words to generate a corpus, and converting the corpus words into word vectors according to the GloVe model, filling missing values, deleting outliers and standardizing the collected numerical data; The S3 pricing model construction is used to import the processed data into the mathematical model of enterprise scale coefficient, the mathematical model of industry risk coefficient, the mathematical model of equipment operation health coefficient, the mathematical model of safety management system perfection coefficient, the mathematical model of employee safety quality index, the mathematical model of enterprise past accident loss coefficient, the mathematical model of geographical risk index and the mathematical model of insurance claim history scoring coefficient, and finally construct the enterprise premium pricing model; The S4 intelligent insurance recommendation: recommends a safety liability insurance product package that meets the pricing for the enterprise based on the enterprise's safety liability insurance premium pricing, and selects a high-priority safety liability insurance product package based on the enterprise's preference priority list; The S5 automated policy management: generates electronic policies based on the insurance information of the selected occupational liability insurance product package. Enterprises can view, download and print the policies online at any time. At the same time, a policy expiration reminder function is set to send renewal notices to the enterprise and insurance company a certain time in advance, and automatically match occupational liability insurance product packages according to changes in the enterprise's pricing model.

[0007] Preferably, the mathematical model of the enterprise scale coefficient is as follows: ,in refers to the enterprise scale coefficient, Refers to the annual turnover of an enterprise. Refers to the total number of employees in an enterprise; The mathematical model of the industry risk coefficient is as follows: +0.2 ,in Refers to the industry risk factor, Refers to the risk coefficient of the basic industry, Refers to the average asset loss rate of an enterprise due to accidents. Refers to the average accident asset loss rate of the industry in which the enterprise is located; The mathematical model of the equipment operation health coefficient is as follows: 0.2 EDP MCP, where Refers to the equipment operation health coefficient. Refers to the equipment failure rate, EDP ​​refers to the proportion of abnormal equipment operation time, and MCP refers to the proportion of equipment maintenance costs; The mathematical model of the safety management system perfection coefficient is as follows: ,in Refers to the safety management system perfection coefficient, Refers to the integrity score of the enterprise's safety system. Refers to the frequency score of enterprise safety system updates. Refers to the score of the enterprise's safety system implementation and supervision; The mathematical model of the employee safety quality index is as follows: ,in Refers to the employee safety quality index, Refers to the proportion of employees with bachelor degree or above. Refers to the average length of safety training per person. Refers to the industry's minimum safety training duration requirements. Refers to the proportion of human factors in past accidents; The mathematical model of the enterprise's past accident loss coefficient is as follows: ,in Refers to the enterprise's past accident loss coefficient, Refers to the total loss amount of the enterprise in the past five years due to accidents. Refers to the average annual output value of the enterprise during the same period; The mathematical model of the geographic risk index is as follows: ,in Refers to the Geographic Risk Index, Refers to the frequency of a certain natural disaster occurring in the location of the enterprise. It refers to the average intensity of a natural disaster, and n refers to the type of natural disaster; The mathematical model of the insurance claim history scoring coefficient is as follows: ,in Refers to the insurance claim history score coefficient. Refers to the number of historical claims. The amount of the claim. Refers to the claim fraud suspicion coefficient.

[0008] Preferably, the specific processing method of the GloVe model is as follows: The preprocessed text data is integrated according to the classification to form a unified corpus. The text in the corpus is scanned and the sentences in the corpus are traversed sentence by sentence. The number of co-occurrences of each word with other words in the window range of 5 words on the left and right is counted to construct a co-occurrence matrix. The size of the co-occurrence matrix is ​​N×N, where N refers to the number of words in the corpus. The word vector and bias term are initialized by randomly taking values ​​in a normal distribution within the range of , and the word vector and bias term are updated according to the minimization of the loss function. In each iteration, the predicted co-occurrence probability is calculated according to the current word vector and bias term and compared with the actual co-occurrence probability obtained from the co-occurrence matrix, the loss function is calculated, and the parameters are updated by backpropagation according to the loss function to make the predicted co-occurrence probability closer to the actual probability, until the decrease of the loss function is less than 0.0001.

[0009] Preferably, the method for obtaining the basic industry risk coefficient is as follows: Classify different industries as high-risk industries =4. Medium and high risk industries =3. Medium and low risk industries =2 and low risk industries =1, high-risk industries refer to industries with an average of 5 or more deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, medium-high-risk industries refer to industries with an average of 2-5 deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, medium-low-risk industries refer to industries with an average of 1-2 deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, and low-risk industries refer to industries with an average of less than 1 death or property loss of more than 100,000 yuan per 100 enterprises per year in the past five years.

[0010] Preferably, the integrity score of the enterprise safety system specifically refers to: matching the word vector extracted from the enterprise safety management system with the word vector extracted from the industry safety standard specification through a cosine similarity algorithm based on the word vector, a cosine similarity greater than 0.8 indicates the same semantics, and counting the proportion of the clauses in the enterprise safety management system that are semantically identical to the clauses in the industry safety standard specification, multiplying the proportion by 10 and rounding it off to the nearest integer, which is the integrity score of the enterprise safety system; The cosine similarity algorithm refers to = , , are two text vectors, is the vector dot product, is the magnitude of the vector.

[0011] Preferably, the enterprise safety system update frequency score specifically refers to calculating the number of version updates of the safety system document within one year. For example, by querying the log records of the version management system, the number of times the version number of the safety system document changes from January 1, 2023 to December 31, 2023 is counted. If the version number is updated from 1.0 to 1.3, the number of updates is 3 times. The update frequency is calculated according to the statistical period. For example, if it is updated 3 times a year, the update frequency is 3 times / year. The enterprise safety system update frequency score is calculated based on the average update frequency of the industry. For example, if the update frequency is 3 times / year and the average update frequency of the industry is 5 times / year, the score can be calculated as (3 / 5) × 10 = 6 points.

[0012] Preferably, the score for the supervision on the implementation of enterprise safety systems specifically refers to: calculating the on-time completion rate of safety inspections, that is, the ratio of the number of safety inspections completed on time to the total number of inspections, as well as the timeliness rate of rectification after hidden dangers are discovered. The score for the supervision on the implementation of enterprise safety systems = (0.6 × on-time completion rate of safety inspections + 0.4 × timeliness rate of rectification after hidden dangers are discovered) × 10.

[0013] Preferably, the proportion of human factors in past accidents specifically refers to: constructing a human factor keyword dictionary, the keyword dictionary includes words such as operating errors, illegal behaviors, fatigue work, and lack of concentration, and counting the proportion of accidents in which the words in the keyword dictionary appear in the company's past accident reports.

[0014] Preferably, the enterprise premium pricing model is as follows: ,in Refers to the pricing of liability insurance premiums. The basic pricing coefficient is the company's annual turnover × 0.2%. refers to the enterprise scale coefficient, Refers to the industry risk factor, Refers to the equipment operation health coefficient. Refers to the safety management system perfection coefficient, Refers to the employee safety quality index, Refers to the enterprise's past accident loss coefficient, Refers to the Geographic Risk Index, Refers to the insurance claims history scoring coefficient.

[0015] Preferably, the enterprise preference priority list specifically refers to a priority ranking table set according to the enterprise preferences. The enterprise sets the priority ranking table according to conditions such as the compensation scope, payment period, accident type coverage, and single accident compensation limit of the occupational liability insurance product package, and gives priority to occupational liability insurance product packages with higher quality based on high priority conditions.

[0016] Preferably, an intelligent occupational liability insurance information management system comprises: a system central processing unit module, a system operation database, a user information terminal, a multi-source data collection module, a multi-source data processing and mining module, an occupational liability insurance pricing data analysis module, an intelligent insurance recommendation module and an automated policy management module; The system operation database includes all data texts of an intelligent occupational liability insurance information management system, and collects information texts output by each module in real time. The system central processor module is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving an intelligent occupational liability insurance information management system; The multi-source data acquisition module is used to deploy sensors in the production equipment and working environment of the enterprise, collect equipment operation data, environmental parameters and employee operation behavior data in the production process in real time, connect with the database of the safety supervision department, obtain the enterprise's safety inspection records, administrative penalty information, industry safety standards and specifications data, integrate the enterprise's internal safety management system data, including safety management system documents, safety training records, hidden danger investigation records and rectification ledger records, collect safety production accident information in the industry through web crawler technology, and output the obtained data to the multi-source data processing and mining module; The multi-source data processing and mining module is used to pre-process the collected multi-source data, clean the collected text data, remove numbers, dates, advertisements, emoticons, hyperlinks and modal particles, extract words from sentences according to the dictionary and classify the words to generate a corpus, and convert the corpus words into word vectors according to the GloVe model, fill missing values, delete outliers and standardize the collected numerical data, and output the processed data to the occupational liability insurance pricing data analysis module; The occupational liability insurance pricing data analysis module is used to import the processed data into the mathematical model of enterprise scale coefficient, the mathematical model of industry risk coefficient, the mathematical model of equipment operation health coefficient, the mathematical model of safety management system perfection coefficient, the mathematical model of employee safety quality index, the mathematical model of enterprise past accident loss coefficient, the mathematical model of geographical risk index, the mathematical model of insurance claims history scoring coefficient and the enterprise premium pricing model to obtain the enterprise occupational liability insurance premium pricing; The intelligent insurance recommendation module is used to recommend a safety liability insurance product package that meets the pricing for the enterprise according to the enterprise safety liability insurance premium pricing, and select a high-priority safety liability insurance product package according to the enterprise preference priority list; The automated insurance policy management module is used to generate an electronic insurance policy based on the insurance information of the selected occupational liability insurance product package. The enterprise can view, download and print the insurance policy online at any time. At the same time, a policy expiration reminder function is set to send a renewal notice to the enterprise and the insurance company a certain time in advance, and automatically match the occupational liability insurance product package according to changes in the enterprise's pricing model.

[0017] Technical effects and advantages of the present invention: 1. The present invention uses a variety of data sources and advanced data analysis methods to build a mathematical model. It uses natural language processing technology to deeply analyze enterprise-related text data. It also combines multi-dimensional factors such as enterprise scale, industry risk coefficient, equipment operation health, and safety management system perfection. With the continuous input of new data, the model can automatically learn and optimize, and can track enterprise risk changes in real time and dynamically, providing risk levels and probability assessments that are more in line with actual conditions; 2. The present invention generates a corpus and constructs a co-occurrence matrix by converting the corpus vocabulary into word vectors according to the GloVe model. This co-occurrence matrix can reflect the semantic and grammatical relationship between words, so that the word vector can better reflect the role of the vocabulary in the entire language environment, thereby better analyzing text data; 3. The present invention fully considers the characteristics and risk factors of the industry during the data preprocessing process, such as adjusting the correlation coefficients and indicators according to the risk rating data, average accident loss rate, etc. of different industries, so that the preprocessed data can better reflect the actual situation and risk level of the industry, and improves the adaptability and accuracy of the model in different industry applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the overall structure of the present invention; Figure 2 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] As attached Figure 1 An intelligent method for the management of occupational liability insurance information is shown, comprising S1: multi-source data fusion collection, S2: multi-source data processing and mining, S3: pricing model construction, S4: intelligent insurance recommendation and S5: automated policy management.

[0021] The S1 multi-source data fusion collection: using the Internet of Things technology, deploying sensors in the enterprise's production equipment and working environment, collecting equipment operation data, environmental parameters and employee operation behavior data in the production process in real time, connecting with the database of the safety supervision department, obtaining the enterprise's safety inspection records, administrative penalty information, industry safety standards and specifications data, integrating the enterprise's internal safety management system data, including safety management system documents, safety training records, hidden danger investigation records and rectification ledger records, and collecting safety production accident information in the industry through web crawler technology; The S2 multi-source data processing and mining: pre-processing the collected multi-source data, cleaning the collected text data, removing numbers, dates, advertisements, emoticons, hyperlinks and modal particles, extracting words from sentences according to the dictionary and classifying the words to generate a corpus, and converting the corpus words into word vectors according to the GloVe model, filling missing values, deleting outliers and standardizing the collected numerical data; The specific processing method of the GloVe model is as follows: The preprocessed text data is integrated according to the classification to form a unified corpus. The text in the corpus is scanned and the sentences in the corpus are traversed sentence by sentence. The number of co-occurrences of each word with other words in the window range of 5 words on the left and right is counted to construct a co-occurrence matrix. The size of the co-occurrence matrix is ​​N×N, where N refers to the number of words in the corpus. Randomly select values ​​in a normal distribution within the range to initialize the word vector and bias term, and update the word vector and bias term according to the minimization loss function. In each iteration, the predicted co-occurrence probability is calculated based on the current word vector and bias term and compared with the actual co-occurrence probability obtained from the co-occurrence matrix, and the loss is calculated. The parameters are updated according to the loss backpropagation to make the predicted co-occurrence probability closer to the actual probability, until the decrease of the loss function is less than 0.0001; The S3 pricing model construction is used to import the processed data into the mathematical model of enterprise scale coefficient, the mathematical model of industry risk coefficient, the mathematical model of equipment operation health coefficient, the mathematical model of safety management system perfection coefficient, the mathematical model of employee safety quality index, the mathematical model of enterprise past accident loss coefficient, the mathematical model of geographical risk index and the mathematical model of insurance claim history scoring coefficient, and finally construct the enterprise premium pricing model; The mathematical model of the enterprise scale coefficient is as follows: ,in refers to the enterprise scale coefficient, Refers to the annual turnover of an enterprise. Refers to the total number of employees in an enterprise; The mathematical model of the industry risk coefficient is as follows: +0.2 ,in Refers to the industry risk factor, Refers to the risk coefficient of the basic industry, Refers to the average asset loss rate of an enterprise due to accidents. Refers to the average accident asset loss rate of the industry in which the enterprise is located; It should be specifically explained in this embodiment that the method for obtaining the basic industry risk coefficient is as follows: Classify different industries as high-risk industries =4. Medium and high risk industries =3. Medium and low risk industries =2 and low risk industries =1, high-risk industries refer to industries with an average of 5 or more deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, medium-high-risk industries refer to industries with an average of 2-5 deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, medium-low-risk industries refer to industries with an average of 1-2 deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, and low-risk industries refer to industries with an average of less than 1 death or property loss of more than 100,000 yuan per 100 enterprises per year in the past five years; The mathematical model of the equipment operation health coefficient is as follows: 0.2 EDP MCP, where Refers to the equipment operation health coefficient. Refers to the equipment failure rate, EDP ​​refers to the proportion of abnormal equipment operation time, and MCP refers to the proportion of equipment maintenance costs; The mathematical model of the safety management system perfection coefficient is as follows: ,in Refers to the safety management system perfection coefficient, Refers to the integrity score of the enterprise's safety system. Refers to the frequency score of enterprise safety system updates. Refers to the score of the enterprise's safety system implementation and supervision; It should be specifically explained in this embodiment that the integrity score of the enterprise safety system specifically refers to: matching the word vector extracted from the enterprise safety management system with the word vector extracted from the industry safety standard specification through the cosine similarity algorithm based on the word vector, and the cosine similarity greater than 0.8 indicates the same semantics, and counting the proportion of the clauses in the enterprise safety management system that are semantically identical to the clauses in the industry safety standard specification, and the proportion is multiplied by 10 and rounded to the nearest integer, which is the integrity score of the enterprise safety system; The cosine similarity algorithm refers to = , , are two text vectors, is the vector dot product, is the magnitude of the vector; The enterprise safety system update frequency score specifically refers to calculating the number of version updates of the safety system document within a year. For example, by querying the log records of the version management system, the number of changes in the version number of the safety system document from January 1, 2023 to December 31, 2023 is counted. If the version number is updated from 1.0 to 1.3, the number of updates is 3 times. The update frequency is calculated based on the statistical period. For example, if it is updated 3 times a year, the update frequency is 3 times / year. The enterprise safety system update frequency score is calculated based on the average update frequency of the industry. If the update frequency is 3 times / year and the average update frequency of the industry is 5 times / year, the score can be calculated as (3 / 5)×10=6 points. The score of enterprise safety system implementation supervision specifically refers to: calculating the on-time completion rate of safety inspections, that is, the ratio of the number of safety inspections completed on time to the total number of inspections, and the timely rectification rate after the hidden dangers are discovered. The score of enterprise safety system implementation supervision = (0.6 × on-time completion rate of safety inspections + 0.4 × timely rectification rate after the hidden dangers are discovered) × 10; The mathematical model of the employee safety quality index is as follows: ,in Refers to the employee safety quality index, Refers to the proportion of employees with bachelor degree or above. Refers to the average length of safety training per person. Refers to the industry's minimum safety training duration requirements. Refers to the proportion of human factors in past accidents; It should be specifically explained in this embodiment that the proportion of human factors in past accidents specifically refers to: constructing a human factor keyword dictionary, the keyword dictionary includes the words of operation error, violation behavior, fatigue operation, and lack of concentration, and counting the proportion of accidents in which the words in the keyword dictionary appear in the past accident reports of the enterprise; The mathematical model of the enterprise's past accident loss coefficient is as follows: ,in Refers to the enterprise's past accident loss coefficient, Refers to the total loss amount of the enterprise in the past five years due to accidents. Refers to the average annual output value of the enterprise during the same period; The mathematical model of the geographic risk index is as follows: ,in Refers to the Geographic Risk Index, Refers to the frequency of a certain natural disaster occurring in the location of the enterprise. It refers to the average intensity of a natural disaster, and n refers to the type of natural disaster; The mathematical model of the insurance claim history scoring coefficient is as follows: ,in Refers to the insurance claim history score coefficient. Refers to the number of historical claims. The amount of the claim. Refers to the claim fraud suspicion coefficient; The claim fraud suspicion coefficient is determined based on the insurance company's internal risk assessment, with suspicion being 1 and no suspicion being 0; The enterprise premium pricing model is as follows: ,in Refers to the pricing of liability insurance premiums. The basic pricing coefficient is the company's annual turnover × 0.2%. refers to the enterprise scale coefficient, Refers to the industry risk factor, Refers to the equipment operation health coefficient. Refers to the safety management system perfection coefficient, Refers to the employee safety quality index, Refers to the enterprise's past accident loss coefficient, Refers to the Geographic Risk Index, Refers to the insurance claims history scoring coefficient; The S4 intelligent insurance recommendation: recommends a safety liability insurance product package that meets the pricing for the enterprise based on the enterprise's safety liability insurance premium pricing, and selects a high-priority safety liability insurance product package based on the enterprise's preference priority list; It should be specifically explained in this embodiment that the enterprise preference priority list specifically refers to a priority ranking table set according to the enterprise preference. The enterprise sets the priority ranking table according to the compensation scope, payment cycle, accident type coverage, single accident compensation limit and other conditions of the occupational liability insurance product package, and gives priority to the occupational liability insurance product package with higher quality in terms of high priority conditions; The S5 automated policy management: generates electronic policies based on the selected occupational liability insurance product package insurance information, and the enterprise can view, download and print the policies online at any time. At the same time, the policy expiration reminder function is set to send renewal notices to the enterprise and insurance company in advance, and automatically matches occupational liability insurance product packages according to changes in the enterprise's pricing model; As attached Figure 2 An intelligent occupational liability insurance information management system shown includes: a system central processing unit module, a system operation database, a user information terminal, a multi-source data collection module, a multi-source data processing and mining module, an occupational liability insurance pricing data analysis module, an intelligent insurance recommendation module and an automated policy management module; The system operation database includes all data texts of an intelligent occupational liability insurance information management system, and collects information texts output by each module in real time. The system central processor module is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving an intelligent occupational liability insurance information management system; The multi-source data acquisition module is used to deploy sensors in the production equipment and working environment of the enterprise, collect equipment operation data, environmental parameters and employee operation behavior data in the production process in real time, connect with the database of the safety supervision department, obtain the enterprise's safety inspection records, administrative penalty information, industry safety standards and specifications data, integrate the enterprise's internal safety management system data, including safety management system documents, safety training records, hidden danger investigation records and rectification ledger records, collect safety production accident information in the industry through web crawler technology, and output the obtained data to the multi-source data processing and mining module; The multi-source data processing and mining module is used to pre-process the collected multi-source data, clean the collected text data, remove numbers, dates, advertisements, emoticons, hyperlinks and modal particles, extract words from sentences according to the dictionary and classify the words to generate a corpus, and convert the corpus words into word vectors according to the GloVe model, fill missing values, delete outliers and standardize the collected numerical data, and output the processed data to the occupational liability insurance pricing data analysis module; The occupational liability insurance pricing data analysis module is used to import the processed data into the mathematical model of enterprise scale coefficient, the mathematical model of industry risk coefficient, the mathematical model of equipment operation health coefficient, the mathematical model of safety management system perfection coefficient, the mathematical model of employee safety quality index, the mathematical model of enterprise past accident loss coefficient, the mathematical model of geographical risk index, the mathematical model of insurance claims history scoring coefficient and the enterprise premium pricing model to obtain the enterprise occupational liability insurance premium pricing; The intelligent insurance recommendation module is used to recommend a safety liability insurance product package that meets the pricing for the enterprise according to the enterprise safety liability insurance premium pricing, and select a high-priority safety liability insurance product package according to the enterprise preference priority list; The automated insurance policy management module is used to generate an electronic insurance policy based on the insurance information of the selected occupational liability insurance product package. The enterprise can view, download and print the insurance policy online at any time. At the same time, a policy expiration reminder function is set to send a renewal notice to the enterprise and the insurance company a certain time in advance, and automatically match the occupational liability insurance product package according to changes in the enterprise's pricing model.

[0022] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other; Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent method for managing occupational liability insurance information, characterized in that: include: S1: Multi-source data fusion collection, S2: Multi-source data processing and mining, S3: Pricing model construction, S4: Intelligent insurance recommendation and S5: Automated policy management; The S1 multi-source data fusion collection: using the Internet of Things technology, deploying sensors in the enterprise's production equipment and working environment, collecting equipment operation data, environmental parameters and employee operation behavior data in the production process in real time, connecting with the database of the safety supervision department, obtaining the enterprise's safety inspection records, administrative penalty information, industry safety standards and specifications data, integrating the enterprise's internal safety management system data, including safety management system documents, safety training records, hidden danger investigation records and rectification ledger records, and collecting safety production accident information in the industry through web crawler technology; The S2 multi-source data processing and mining: pre-processing the collected multi-source data, cleaning the collected text data, removing numbers, dates, advertisements, emoticons, hyperlinks and modal particles, extracting words from sentences according to the dictionary and classifying the words to generate a corpus, and converting the corpus words into word vectors according to the GloVe model, filling missing values, deleting outliers and standardizing the collected numerical data; The S3 pricing model construction is used to import the processed data into the mathematical model of enterprise scale coefficient, the mathematical model of industry risk coefficient, the mathematical model of equipment operation health coefficient, the mathematical model of safety management system perfection coefficient, the mathematical model of employee safety quality index, the mathematical model of enterprise past accident loss coefficient, the mathematical model of geographical risk index and the mathematical model of insurance claim history scoring coefficient, and finally construct the enterprise premium pricing model; The S4 intelligent insurance recommendation: recommends a safety liability insurance product package that meets the pricing for the enterprise based on the enterprise's safety liability insurance premium pricing, and selects a high-priority safety liability insurance product package based on the enterprise's preference priority list; The S5 automated policy management: generates electronic policies based on the insurance information of the selected occupational liability insurance product package. Enterprises can view, download and print the policies online at any time. At the same time, a policy expiration reminder function is set to send renewal notices to the enterprise and insurance company a certain time in advance, and automatically match occupational liability insurance product packages according to changes in the enterprise's pricing model.

2. The intelligent method for managing occupational liability insurance information according to claim 1, characterized in that: The mathematical model of the enterprise scale coefficient is as follows: ,in refers to the enterprise scale coefficient, Refers to the annual turnover of an enterprise. Refers to the total number of employees in an enterprise; The mathematical model of the industry risk coefficient is as follows: +0.2 ,in Refers to the industry risk factor, Refers to the risk coefficient of the basic industry, Refers to the average asset loss rate of an enterprise due to accidents. Refers to the average accident asset loss rate of the industry in which the enterprise is located; The mathematical model of the equipment operation health coefficient is as follows: 0.2 EDP MCP, where Refers to the equipment operation health coefficient, Refers to the equipment failure rate, EDP ​​refers to the proportion of abnormal equipment operation time, and MCP refers to the proportion of equipment maintenance costs; The mathematical model of the safety management system perfection coefficient is as follows: ,in Refers to the safety management system perfection coefficient, Refers to the integrity score of the enterprise's safety system. Refers to the frequency score of enterprise safety system updates. Refers to the score of the enterprise's safety system implementation and supervision; The mathematical model of the employee safety quality index is as follows: ,in Refers to the employee safety quality index, Refers to the proportion of employees with bachelor degree or above. Refers to the average length of safety training per person. Refers to the industry's minimum safety training duration requirements. Refers to the proportion of human factors in past accidents; The mathematical model of the enterprise's past accident loss coefficient is as follows: ,in Refers to the enterprise's past accident loss coefficient, Refers to the total amount of losses caused by accidents in the past five years. Refers to the average annual output value of the enterprise during the same period; The mathematical model of the geographic risk index is as follows: ,in Refers to the Geographic Risk Index, Refers to the frequency of a certain natural disaster occurring in the location of the enterprise. It refers to the average intensity of a natural disaster, and n refers to the type of natural disaster; The mathematical model of the insurance claim history scoring coefficient is as follows: ,in Refers to the insurance claim history score coefficient, Refers to the number of historical claims. The amount of the claim. Refers to the claim fraud suspicion coefficient.

3. The intelligent method for managing occupational liability insurance information according to claim 2, characterized in that: The method for obtaining the basic industry risk coefficient is as follows: Classify different industries as high-risk industries =4. Medium and high risk industries =3. Medium and low risk industries =2 and low risk industries =1, high-risk industries refer to industries with an average of 5 or more deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, medium-high-risk industries refer to industries with an average of 2-5 deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, medium-low-risk industries refer to industries with an average of 1-2 deaths or property losses of more than 100,000 yuan per 100 enterprises per year in the past five years, and low-risk industries refer to industries with an average of less than 1 death or property loss of more than 100,000 yuan per 100 enterprises per year in the past five years.

4. The intelligent method for managing occupational liability insurance information according to claim 2 is characterized by: The integrity score of the enterprise safety system specifically refers to: matching the word vector extracted from the enterprise safety management system with the word vector extracted from the industry safety standards and specifications through the cosine similarity algorithm based on the word vector, and the cosine similarity greater than 0.8 indicates the same semantics, and counting the proportion of the clauses in the enterprise safety management system that are semantically identical to the clauses in the industry safety standards and specifications. The proportion is multiplied by 10 and rounded to the nearest integer, which is the integrity score of the enterprise safety system; The cosine similarity algorithm refers to = , , are two text vectors, is the vector dot product, is the magnitude of the vector; The enterprise safety system update frequency score specifically refers to calculating the number of version updates of the safety system document within a year. For example, by querying the log records of the version management system, the number of changes in the version number of the safety system document from January 1, 2023 to December 31, 2023 is counted. If the version number is updated from 1.0 to 1.3, the number of updates is 3 times. The update frequency is calculated based on the statistical period. For example, if it is updated 3 times a year, the update frequency is 3 times / year. The enterprise safety system update frequency score is calculated based on the average update frequency of the industry. If the update frequency is 3 times / year and the average update frequency of the industry is 5 times / year, the score can be calculated as (3 / 5)×10=6 points. The score for the supervision of the implementation of enterprise safety systems specifically refers to: calculating the on-time completion rate of safety inspections, that is, the ratio of the number of safety inspections completed on time to the total number of inspections, as well as the timeliness rate of rectification after hidden dangers are discovered. The score for the supervision of the implementation of enterprise safety systems = (0.6×on-time completion rate of safety inspections + 0.4×timeliness rate of rectification after hidden dangers are discovered)×10.

5. The intelligent method for managing occupational liability insurance information according to claim 1, characterized in that: The proportion of human factors in past accidents specifically refers to: constructing a human factor keyword dictionary, the keyword dictionary includes operating errors, illegal behaviors, fatigue work, and lack of concentration, and counting the proportion of accidents in the company's past accident reports that appeared in the keyword dictionary.

6. The intelligent method for managing occupational liability insurance information according to claim 1, characterized in that: The enterprise premium pricing model is as follows: ,in Refers to the pricing of liability insurance premiums. The basic pricing coefficient is the company's annual turnover × 0.2%. refers to the enterprise scale coefficient, Refers to the industry risk factor, Refers to the equipment operation health coefficient, Refers to the safety management system perfection coefficient, Refers to the employee safety quality index, Refers to the enterprise's past accident loss coefficient, Refers to the Geographic Risk Index, Refers to the insurance claims history scoring coefficient.

7. The intelligent method for managing occupational liability insurance information according to claim 1, characterized in that: The enterprise preference priority list specifically refers to a priority ranking list set according to the enterprise's preferences. The enterprise sets a priority ranking list based on conditions such as the compensation scope, payment period, accident type coverage, and single accident compensation limit of the occupational liability insurance product package, and gives priority to occupational liability insurance product packages with higher quality based on high priority conditions.

8. An intelligent occupational liability insurance information management system, characterized by: System CPU module, system operation database, user information terminal, multi-source data collection module, multi-source data processing and mining module, liability insurance pricing data analysis module, intelligent insurance recommendation module and automated policy management module; The system operation database includes all data texts of an intelligent occupational liability insurance information management system, and collects information texts output by each module in real time. The system central processor module is used to centrally control the information text instructions output by each module. The user information terminal is an information output device for receiving an intelligent occupational liability insurance information management system; The multi-source data acquisition module is used to deploy sensors in the production equipment and working environment of the enterprise, collect equipment operation data, environmental parameters and employee operation behavior data in the production process in real time, connect with the database of the safety supervision department, obtain the enterprise's safety inspection records, administrative penalty information, industry safety standards and specifications data, integrate the enterprise's internal safety management system data, including safety management system documents, safety training records, hidden danger investigation records and rectification ledger records, collect safety production accident information in the industry through web crawler technology, and output the obtained data to the multi-source data processing and mining module; The multi-source data processing and mining module is used to pre-process the collected multi-source data, clean the collected text data, remove numbers, dates, advertisements, emoticons, hyperlinks and modal particles, extract words from sentences according to the dictionary and classify the words to generate a corpus, and convert the corpus words into word vectors according to the GloVe model, fill missing values, delete outliers and standardize the collected numerical data, and output the processed data to the occupational liability insurance pricing data analysis module; The occupational liability insurance pricing data analysis module is used to import the processed data into the mathematical model of enterprise scale coefficient, the mathematical model of industry risk coefficient, the mathematical model of equipment operation health coefficient, the mathematical model of safety management system perfection coefficient, the mathematical model of employee safety quality index, the mathematical model of enterprise past accident loss coefficient, the mathematical model of geographical risk index, the mathematical model of insurance claims history scoring coefficient and the enterprise premium pricing model to obtain the enterprise occupational liability insurance premium pricing; The intelligent insurance recommendation module is used to recommend a safety liability insurance product package that meets the pricing for the enterprise according to the enterprise safety liability insurance premium pricing, and select a high-priority safety liability insurance product package according to the enterprise preference priority list; The automated insurance policy management module is used to generate an electronic insurance policy based on the insurance information of the selected occupational liability insurance product package. The enterprise can view, download and print the insurance policy online at any time. At the same time, a policy expiration reminder function is set to send a renewal notice to the enterprise and the insurance company a certain time in advance, and automatically match the occupational liability insurance product package according to changes in the enterprise's pricing model.

Images