An Insurance Information Management System Based on Neural Network Model

Abstract

The present invention relates to the field of information management technology for specific computing models, and in particular to an insurance information management system based on a neural network model, including a login management module for performing type identification and identity authentication of points, and also for performing data sharding storage and information isolation judgment; an association management module for performing point association to obtain the first point and sequential points of each sub-point; an execution management module for managing the service execution of the first point and sequential points; a point group supervision module for judging the point group supervision situation according to the number of emergency point optimizations and the number of point groups to be responded to, and optimizing the number of emergency point groups and the number of point groups to be responded to according to the judgment result; an information management module for managing the information transmission between sub-points, first points and sequential points. The present invention solves the problems of performance bottlenecks, data islands and low management efficiency, shortens the claims processing time and improves work efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of information management of specific computing models, and in particular to an insurance information management system based on a neural network model. Background Art

[0002] With the rapid development of information technology, insurance information management systems play a vital role in the modern insurance industry. At present, insurance information management systems mainly rely on centralized databases and rule-based algorithms to process large amounts of customer information, policy data and claims records. However, with the continuous increase in data volume and complexity, traditional systems are facing performance bottlenecks, data silos and inefficient management. The advantages of applying neural network models in insurance information management systems are mainly reflected in the fact that neural networks can process and analyze large amounts of complex multidimensional data. By training deep learning models, the system can identify potential fraud. Neural networks can provide personalized insurance product recommendations and services based on customers' historical behaviors and preferences. Through image recognition and natural language processing technology, neural network models can automatically process documents and pictures in claims applications, greatly shortening claims processing time and improving work efficiency.

[0003] Chinese Patent Publication No.: CN111340402A discloses a group insurance service management method, system, device and storage medium, the method comprising: obtaining group member information in a group point from a point information management system; assigning a service group to each group member in the group point, each service group including at least one service personnel; receiving a service personnel image sent by the group member from a point terminal, performing identity authentication to determine whether the service personnel image corresponds to a service personnel in the service group; if so, establishing an association relationship between the group member and the service personnel; sending the identity authentication result to the corresponding point terminal, and when generating a new service order, establishing an association relationship between the service order and the service personnel, and counting service data. However, this solution only adjusts the associated personnel of the group insurance service group, and does not solve problems such as performance bottlenecks, data islands and low management efficiency. The claims processing time is long and the work efficiency is low. Summary of the invention

[0004] To this end, the present invention provides an insurance information management system based on a neural network model to overcome the problems of low management efficiency in the prior art due to performance bottlenecks, data islands and long claims processing time.

[0005] To achieve the above object, the present invention provides an insurance information management system based on a neural network model, comprising:

[0006] The login management module is used to identify the type of the point, to authenticate the identity of the point after the type identification, to store the data of the point that has passed the authentication in slices, to adjust the type identification result, and to perform information isolation judgment on the data slice storage of the point;

[0007] Sharded storage database is used to store the data generated by the points;

[0008] The association management module is used to associate the points that have passed the identity authentication, obtain the first point and the sequential point of each sub-point, optimize the point association process, determine the accuracy of the optimization of the point association process, and adjust the optimization process of the point association process;

[0009] An execution management module is used to manage the service execution of the first point and the sequential point, and to optimize the first point and the sequential point of each sub-point, and to optimize the first point and the sequential point of each sub-point according to the emergency point group;

[0010] The point group supervision module is used to judge the point group supervision situation according to the number of emergency point optimizations and the number of response point groups, and optimize the number of emergency point groups and the number of response point groups according to the judgment results;

[0011] The information management module is used to manage the information transmission between sub-points, primary points and sequential points.

[0012] Furthermore, the login management module obtains the user ID, the user registration identity and the user selected identity, and performs type identification on the point, wherein:

[0013] When the number of user registration identities corresponding to the user ID in the shard storage database is U=1, the login management module uses the user registration identity corresponding to the user ID as the point type;

[0014] When the number of user registered identities corresponding to the user ID in the shard storage database is greater than 1, the login management module pushes an identity selection window to the point and uses the user selected identity as the point type;

[0015] The login management module outputs the point type as a type identification result.

[0016] Furthermore, the login management module obtains the input password, and compares the input password with the preset password corresponding to the user ID in the shard storage database, and performs identity authentication on the point after type identification according to the comparison result, wherein:

[0017] When the input password is consistent with the preset password, the login management module determines that the point after the type identification has passed the identity authentication, and marks it as a point that has passed the identity authentication;

[0018] When the input password is inconsistent with the preset password, the login management module determines that the point after the type recognition has not passed the identity authentication.

[0019] Furthermore, the login management module stores the generated data in a target shard of the shard storage database.

[0020] Furthermore, the login management module compares the point switching type option with the original point type, and determines the adjustment of the type recognition result according to the comparison result, wherein:

[0021] When the point switching type option is consistent with the original point type, the login management module determines not to adjust the type recognition result;

[0022] When the point switching type option is inconsistent with the original point type, the login management module determines to adjust the type recognition result and adjusts the type recognition result to the point switching type option.

[0023] Furthermore, after adjusting the type identification result, the login management module shares information between the target slice of the origin type and the target slice of the adjusted type identification result, and performs relationship analysis on the user to obtain the user ID that has a vested interest in the user, uses it as the relationship ID, and isolates information between the target slice of the relationship ID and the target slice of the user. The login management module repeats information sharing and information isolation for the user.

[0024] Furthermore, when the association management module associates the points that have passed the identity authentication, it generates a recommended portrait of each sub-point in the point group, a reverse portrait of each corresponding point in the corresponding point group, and a reverse portrait of each sequence point in each sequence point group, calculates the first point overlap D1 of the recommended portrait of the sub-point and the reverse portrait of each corresponding point, and the sequence point overlap D2 of the recommended portrait of the sub-point and the reverse portrait of each sequence point, and the association management module associates the corresponding point with the maximum first point overlap D1max as the first point of the sub-point, associates the sequence point with the maximum sequence point overlap D2max among the free sequence points of each sequence as the matching sequence point of the sub-point in each sequence, and connects the matching sequence points of each sequence in series to obtain the sequence points of the sub-point;

[0025] The association management module compares the number of idle points n, and determines the optimization of the point association process according to the comparison result, wherein:

[0026] When n>1, the association management module determines not to optimize the point association process;

[0027] When n=1, the association management module determines not to optimize the point association process;

[0028] When n=0, the association management module determines to optimize the point association process, and uses the earliest released point as the matching point of the sub-point and associates it with the sub-point.

[0029] Furthermore, the association management module compares the association discrimination Q with the preset association discrimination Q0, and makes an accuracy judgment on the optimization of the point association process according to the comparison result, wherein:

[0030] When Q≥Q0, the association management module determines that the optimization of the point association process is accurate, and outputs it as the accuracy judgment result;

[0031] When Q<Q0, the association management module determines that the optimization of the point association process is inaccurate, and outputs it as the accuracy judgment result;

[0032] The association management module adjusts the optimization process of the point association process according to the accuracy judgment result, wherein:

[0033] When the accuracy judgment result indicates that the optimization of the point association process is accurate, the association management module does not adjust the optimization process of the point association process;

[0034] When the accuracy judgment result is that the optimization situation of the point association process is inaccurate, the association management module adjusts the optimization process of the point association process, and when n>1, the association management module adjusts the judgment result of the optimization situation of not optimizing the point association process to optimizing the point association process, and performs association adjustment on the target point corresponding to the accuracy judgment result that the optimization situation of the point association process is inaccurate;

[0035] When the target point is adjusted for association, the association management module optimizes and iterates the preset statistical weight matrix of the point overlap of the target point until the association discrimination of the target point overlap is greater than or equal to the preset association discrimination, and records the number of optimization iterations Y, compares the number of optimization iterations Y with the preset number of optimization iterations Y0, and corrects the association adjustment according to the comparison result, wherein:

[0036] When Y≥Y0, the association management module corrects the association adjustment, increases the image parameters, and optimizes and iterates the preset statistical weight matrix of the point coincidence of the target points according to the increased image parameters until the association discrimination of the target point coincidence is greater than or equal to the preset association discrimination;

[0037] When Y<Y0, the association management module does not correct the association adjustment.

[0038] Furthermore, the execution management module manages the service execution of the primary point and the sequential points according to the sub-point service process, wherein:

[0039] When the sub-point service process is associated, the execution management module notifies the head point of the sub-point to provide service;

[0040] When the sub-point service process is in the current order to be scheduled, the execution management module does not notify the sub-points in the current order to provide services;

[0041] When the sub-point service process is in progress for the current order, the execution management module notifies the sub-points of the current order to provide services;

[0042] When the sub-point service process is completed for the current order, the execution management module does not notify the sub-points of the current order to provide services;

[0043] The execution management module compares the service point response time T with the preset service point response time T0, and optimizes and determines the first point and the sequence point of each sub-point according to the comparison result, wherein:

[0044] When T≤T0, the execution management module determines that the first point and the sequential point of each sub-point respond in time, and does not optimize the first point and the sequential point of each sub-point;

[0045] When T>T0, the execution management module determines that the first point and the sequential point of each sub-point do not respond in time, and optimizes the first point and the sequential point of each sub-point as the response optimization point;

[0046] When optimizing the response optimization point, the response optimization point is sent to the association management module, the matching response optimization point output by the association management module is obtained, and the matching response optimization point is used as the first point and the sequential point of each of the sub-points;

[0047] The execution management module obtains the service point response time T1 of the matching response optimization point, and compares the service point response time T1 of the matching response optimization point with the preset service point response time T0, and judges the optimization effect of the matching response optimization point according to the comparison result, wherein:

[0048] When T1≤T0, the execution management module determines that the optimization effect of the matching response optimization point is good;

[0049] When T1>T0, the execution management module determines that the optimization effect of the matching response optimization point is poor;

[0050] When the execution management module determines that the optimization effect of the matching response optimization point is poor, the emergency point with the smallest release time is selected from the emergency point group as the emergency optimization point of the first point and the sequential point of each sub-point.

[0051] Furthermore, the point group supervision module obtains the total number of emergency points k1, the number of times each emergency point is used p1, the number of response point groups k2 and the number of times each response point is used p2, calculates the utilization rate r1 of a single emergency point and the utilization rate r2 of a single response point, sets r1=p1 / H1, H1 is the total available time of each emergency point, sets r2=p2 / H2, H2 is the total available time of each response point, and the point group supervision module compares the utilization rate r1 of a single emergency point and the utilization rate r2 of a single response point during the peak period with the preset point utilization rate r0, and identifies the bottleneck point according to the comparison results, wherein:

[0052] When r1≥r0, the point group supervision module determines that the emergency point is a bottleneck point;

[0053] When r1<r0, the point group supervision module determines that the emergency point is not a bottleneck point;

[0054] When r2≥r0, the point group supervision module determines that the point is a bottleneck point;

[0055] When r2<r0, the point group supervision module determines that the point is not a bottleneck point;

[0056] The point group supervision module obtains the bottleneck emergency point quantity ka1 and the bottleneck response point quantity ka2, and calculates the resource utilization rate β, setting β=0.7×ka1 / k1+0.3×ka2 / k2. The point group supervision module compares the resource utilization rate β with the preset resource utilization rate β0, and judges the point group supervision situation according to the comparison result, where:

[0057] When β≤β0, the point group supervision module determines that the point group supervision is normal, and does not optimize the number of emergency point groups and the number of response point groups;

[0058] When β>β0, the point group supervision module determines that the point group supervision situation is abnormal, and optimizes the number of emergency point groups and the number of response point groups. The optimized number of emergency point groups is kb1, and kb1=α×k1 is set. The optimized number of response point groups is kb2, and kb2=α×k2 is set. 1<α≤1.2 is set, and α is the point group number optimization coefficient;

[0059] The information management module sends the sub-point service information to the first point and the sequential points, and shares the first point service information and the sequential point service information to the sub-point terminals.

[0060] Compared with the prior art, the beneficial effect of the present invention is that the system is responsible for point type identification and identity authentication through the login management module, which helps to ensure the legitimacy and accuracy of the points, thereby reducing interference from misoperation and illegal points, and the sharding storage of point data after type identification can optimize data management, making the processing of large amounts of data more efficient, and ensuring data security and privacy protection through information isolation judgment. The system associates points through the association management module to obtain the first point and the sequential point, and optimizes and judges the accuracy of the association process, thereby ensuring that the association relationship between points is reasonable and effective, continuously improving the efficiency of point association, and adapting to dynamic In order to meet the dynamically changing business needs, the system manages the service execution of the primary point and the sequential point through the execution management module, and ensures the continuity and stability of the service through optimization judgment and regulation of the emergency point group. Especially in emergency situations, the system judges the point group supervision situation according to the number of emergency point optimizations and the number of emergency point groups through the point group supervision module, and performs corresponding optimization, thereby improving the overall performance and responsiveness of the point group, ensuring that the system can maintain efficient operation under different load conditions. The system manages the information transmission between the sub-points, the primary point and the sequential points through the information management module, ensuring the accuracy and timeliness of information transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a structural diagram of the insurance information management system based on the neural network model of this embodiment. DETAILED DESCRIPTION

[0062] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0063] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0064] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0065] See also Figure 1 As shown, it is a schematic diagram of the structure of the insurance information management system based on the neural network model of this embodiment, and the system includes:

[0066] The login management module is used to identify the type of the point, to authenticate the identity of the point after the type identification, to store the data of the point that has passed the authentication in slices, to adjust the type identification result, and to perform information isolation judgment on the data slice storage of the point;

[0067] The shard storage database is used to store the data generated by the points. The shard storage database is connected with the login management module, the association management module, the execution management module, the point group supervision module and the information management module;

[0068] The association management module is used to associate the points that have passed the identity authentication, obtain the first point and the sequential point of each sub-point, optimize the point association process, determine the accuracy of the optimization of the point association process, and adjust the optimization process of the point association process. The association management module is connected to the login management module;

[0069] An execution management module is used to manage the service execution of the first point and the sequential point, and is also used to optimize the first point and the sequential point of each sub-point, and is also used to optimize the first point and the sequential point of each sub-point according to the emergency point group. The execution management module is connected with the association management module;

[0070] The point group supervision module is used to judge the point group supervision situation according to the number of emergency point optimizations and the number of response point groups, and optimize the number of emergency point groups and the number of response point groups according to the judgment results. The point group supervision module is connected with the execution management module;

[0071] The information management module is used to manage the information transmission between sub-points, the first point and the sequential points. The information management module is connected to the point group supervision module.

[0072] Specifically, the system is set in the insurance service industry management terminal, and manages insurance information based on a neural network model to avoid performance bottlenecks, data islands, and long claims processing time, thereby improving the efficiency of insurance information management. The system is responsible for point type identification and identity authentication through a login management module, which helps to ensure the legality and accuracy of the points, thereby reducing interference from misoperations and illegal points, and the sharding storage of point data after type identification can optimize data management, making the processing of large amounts of data more efficient, and ensuring data security and privacy protection through information isolation judgment. The system associates points through an association management module to obtain the first point and sequential points, and optimizes and judges the accuracy of the association process, thereby ensuring To ensure that the relationship between points is reasonable and effective, continuously improve the efficiency of point association, and adapt to dynamically changing business needs, the system manages the service execution of the first point and sequential points through the execution management module, and ensures the continuity and stability of the service through optimization judgment and regulation of emergency point groups. Especially in emergency situations, the system judges the point group supervision situation according to the number of emergency point optimizations and the number of emergency point groups through the point group supervision module, and performs corresponding optimization, thereby improving the overall performance and responsiveness of the point group, ensuring that the system can maintain efficient operation under different load conditions. The system manages the information transmission between sub-points, first points and sequential points through the information management module to ensure the accuracy and timeliness of information transmission.

[0073] Specifically, the login management module obtains the user ID, user registration identity and user selected identity, and performs type identification on the point, wherein:

[0074] When the number of user registration identities corresponding to the user ID in the shard storage database is U=1, the login management module uses the user registration identity corresponding to the user ID as the point type;

[0075] When the number of user registered identities corresponding to the user ID in the shard storage database is greater than 1, the login management module pushes an identity selection window to the point and uses the user selected identity as the point type;

[0076] The login management module outputs the point type as a type identification result.

[0077] Specifically, the point refers to the terminal where the user logs in, the user ID refers to a serial number of a combination of numbers and letters used to distinguish the user identity, such as A13h24, the user registration identity refers to the task role corresponding to the responsibilities and rights assumed by the user in the insurance information management system when registering, including the insured, insurance agent, reinsurance company personnel, supervisory agency personnel, legal compliance experts, medical professionals and financial accounting personnel, the user selected identity refers to the user registration identity selected by the user in the identity selection window, the number of user registration identities refers to the total number of task roles corresponding to the responsibilities and rights assumed by the user in the insurance information management system when registering, and the identity selection window refers to an interactive window pushed by the login management module to the point for the user to select the user registration identity at each point. This embodiment does not limit the setting method of the identity selection window, and those skilled in the art can freely set it according to actual conditions, as long as the need to collect user selections is met. For example, the identity selection window can be set to be pushed to the point, with the user registration identity as an option, the option checked by the user is obtained, and the user registration identity checked by the user is used as the user selected identity.

[0078] Specifically, when the number of user registered identities U=1, the login management module directly uses the registered identity as the point type to simplify the operation process and improve the response speed of the system. When the number of user registered identities U>1, the system pushes an identity selection window, and the user can select a specific identity to log in, so that the user can choose a suitable identity according to different scenarios, increasing the flexibility of the system and the friendliness of the user experience. The login management module associates the user's registered identity and the selected identity through the user ID, so that the system can quickly identify the point type, ensuring that the user's operating permissions and responsibilities under different identities are clear, thereby improving the security and accuracy of the system, and managing user data through a sharded storage database, and performing identity authentication and type identification in the login management module, ensuring that the sharded storage and access permission control of the data are strict, and improving data security and privacy protection capabilities.

[0079] Specifically, the login management module obtains the input password, and compares the input password with the preset password corresponding to the user ID in the shard storage database, and performs identity authentication on the point after type recognition according to the comparison result, wherein:

[0080] When the input password is consistent with the preset password, the login management module determines that the point after the type identification has passed the identity authentication, and marks it as a point that has passed the identity authentication;

[0081] When the input password is inconsistent with the preset password, the login management module determines that the point after the type recognition has not passed the identity authentication.

[0082] Specifically, the input password refers to the confidentiality code entered by the user at the terminal where the point is located when logging in. This embodiment does not limit the marking method of the points that have passed the identity authentication. Technical personnel in this field can freely set it according to actual conditions. It only needs to meet the need to distinguish between the points that have passed the identity authentication and the points that have not passed the identity authentication. For example, the name of the point that has passed the identity authentication can be suffixed with the letter A.

[0083] Specifically, the login management module stores the generated data in the target shard of the shard storage database.

[0084] Specifically, the generated data refers to the data generated in real time by the identity-authenticated point, and the target shard of the sharded storage database refers to the partitioned database corresponding to the identity-authenticated point in the sharded storage database, which is used to independently store the data generated in real time by the identity-authenticated point, and the index of the target shard is the user ID.

[0085] Specifically, the sharded storage mechanism of the login management module makes the access and management of data between different shards more efficient, thereby improving the overall data management capabilities of the system, and improving the efficiency of insurance information management while avoiding performance bottlenecks, data silos, and long claims processing time.

[0086] Specifically, the login management module compares the point switching type option with the original point type, and determines the adjustment of the type recognition result according to the comparison result, wherein:

[0087] When the point switching type option is consistent with the original point type, the login management module determines not to adjust the type recognition result;

[0088] When the point switching type option is inconsistent with the original point type, the login management module determines to adjust the type recognition result and adjusts the type recognition result to the point switching type option.

[0089] Specifically, the point switching type option refers to an option for the user to reselect the user selected identity after logging in to the point, and the original point type refers to the point type before the user selected identity is reselected.

[0090] Specifically, after adjusting the type identification result, the login management module shares information with the target slice of the origin type and the target slice of the adjusted type identification result, and performs relationship analysis on the user to obtain the user ID that has a vested interest in the user, and uses it as the relationship ID. The target slice of the relationship ID is informationally isolated from the target slice of the user, and the login management module repeats information sharing and information isolation for the user.

[0091] Specifically, the information sharing refers to integrating the target slice of the origin type and the target slice of the adjusted type recognition result into one target slice. This embodiment does not specifically limit the way of information sharing, and technical personnel in this field can freely set it according to actual conditions, and only need to meet the information sharing requirements of the target slice of the origin type and the target slice of the adjusted type recognition result. For example, it can be set to perform secret-free transmission when querying, downloading, uploading, etc. the target slice of the origin type and the target slice of the adjusted type recognition result. The relationship analysis refers to analyzing the degree of association and association attributes between user IDs. The existence of a conflict of interest refers to the existence of association attributes of conflicting interests between user IDs, and the degree of association reaches a preset degree of association. For example, it is set The preset association degree is set to 85%. When the association degree is 90%, it is determined that the association degree reaches the preset association degree. The information isolation refers to a measure to isolate the target slice of the user from the target slice of the user ID in which the user has an interest. This embodiment does not limit the way of information isolation. Those skilled in the art can freely set it according to actual conditions. It only needs to meet the requirement of information isolation between the target slice of the user and the target slice of the user ID in which the user has an interest. For example, it can be set to set decryption authority on the target slice of the user and the target slice of the user ID in which the user has an interest. When information is transmitted, there is no decryption authority between the target slice of the user and the target slice of the user ID in which the user has an interest.

[0092] Specifically, the login management module performs relationship analysis on the users through a neural network model, and uses user portrait data of user IDs that have a vested interest in the user and user portrait data of user IDs that have no vested interest in the user as model construction data sets, and divides 70% of the model construction data set into a model training data set, and 30% into a model test data set, and inputs the model training data set into the neural network model to train the neural network model to obtain a trained neural network model, and uses the model test data set to test the trained neural network model, and when the proportion of data that is consistent with the vested interest results in the model test data set after the test is greater than a preset data proportion, the trained neural network model is output as a relationship analysis model, and the login management module can input the user portrait data of the user and each user ID into the relationship analysis model, perform relationship analysis on the users, and obtain user IDs that have a vested interest in the users.

[0093] Specifically, the login management module implements information sharing and information isolation mechanisms and uses a neural network model to analyze user relationships, effectively isolating target slices of user IDs in which users have a vested interest, preventing sensitive information from being leaked, and improving the overall data security of the system. The system allows target slices of the origin type to share information with target slices of the adjusted type recognition results, so as to improve the efficiency of information sharing through a non-cryptographic transmission method and reduce the complexity and time cost of data transmission.

[0094] Specifically, the shard storage database stores data generated by sub-points, primary points, sequential points and emergency points.

[0095] It can be understood that the present embodiment does not specifically limit the content of the data generated by the sub-points, first point, sequential points and emergency points. Technical personnel in this field can freely set it according to the actual situation, and only need to meet the shard storage requirements. For example, the service information of the sub-point can be stored in the target shard of the sub-point, the service information of the first point can be stored in the target shard of the first point, and the service information of the sequential points can be stored in the target shard of the sequential points.

[0096] Specifically, when the association management module associates points that have passed identity authentication, it generates recommended portraits of each sub-point in the point group, reverse portraits of each corresponding point in the corresponding point group, and reverse portraits of each sequence point in each sequence point group, calculates the first point overlap D1 between the recommended portraits of the sub-point and the reverse portraits of each corresponding point, and the sequence point overlap D2 between the recommended portraits of the sub-point and the reverse portraits of each sequence point, and the association management module associates the corresponding point with the maximum first point overlap D1max as the first point of the sub-point, associates the sequence point with the maximum sequence point overlap D2max among the free sequence points of each sequence as the matching sequence point of the sub-point in each sequence, and connects the matching sequence points of each sequence in series to obtain the sequence points of the sub-point.

[0097] Specifically, the point group refers to a terminal set where a user with a registered identity as a consumer role of the insured person logs in as a group, the sub-point refers to a terminal where a user with a registered identity as a consumer role of the insured person logs in as a group in the point group, the recommended portrait refers to data reflecting the consumption habits and preferences of the sub-point, the response point group refers to a terminal set where a user with a registered identity as a service provider role of an insurance agent logs in as a group, the response point refers to a terminal where a user with a registered identity as a service provider role of an insurance agent logs in as a group in the response point group, the reverse portrait of the response point refers to data reflecting the service characteristics and quality of the response point, the respective sequence point groups refer to respective sets of terminals where a user with a registered identity as a service provider role of a reinsurance company personnel, a supervisory agency personnel, a legal compliance expert, a medical professional, and a financial accountant personnel logs in as a group classified according to the time sequence of providing services to the sub-points, the sequence point refers to each point in the respective sequence point groups, and the reverse portrait of the sequence point refers to data reflecting the service characteristics and quality of the sequence point. The first point coincidence degree refers to the degree of matching between the recommended portrait of the sub-point and the reverse portrait of each corresponding point, which indicates the degree of consistency between the service characteristics and quality of each corresponding point and the consumption habits and preferences of the sub-point. In this embodiment, the cosine similarity of each parameter in the recommended portrait of the sub-point is calculated with the corresponding parameter in the reverse portrait of each corresponding point to obtain the cosine similarity of each parameter, and the cosine similarity of each parameter is summed up through a preset statistical weight matrix to obtain the first point coincidence degree. Each parameter in the recommended portrait of the sub-point refers to The various values ​​in the data reflecting the consumption habits and preferences of the sub-points, such as the maximum consumption amount, the corresponding parameters in the reverse portrait of each point refer to the various values ​​of the data reflecting the service characteristics and quality of the point, such as the maximum consumption amount of historical customers served, and the preset statistical weight matrix refers to the summation coefficient set according to the importance when the cosine similarities of each parameter are added. For example, when the cosine similarities of 5 parameters are added, the preset statistical weight matrix is ​​set to [0.1, 0.3, 0.05, 0.25, 0.3], the order point coincidence degree refers to the degree of matching between the recommended portrait of the sub-point and the reverse portrait of each order point, which indicates the degree of matching between the service characteristics and quality of each order point and the consumption habits and preferences of the sub-point. In this embodiment, the cosine similarity of each parameter in the recommended portrait of the sub-point and the corresponding parameter in the reverse portrait of each order point is calculated to obtain the cosine similarity of each parameter, and the cosine similarity of each parameter is summed up through a preset statistical weight matrix to obtain the order point coincidence degree. The corresponding parameters in the reverse portrait of each order point refer to the various values ​​of the data reflecting the service characteristics and quality of the order point, such as the maximum consumption amount of historical customers served. The maximum first point coincidence degree D1max refers to the cosine similarity of each parameter. The maximum value of the overlap between the point and the sub-point indicates the maximum degree of fit between the service characteristics and quality of the point and the consumption habits and preferences of the sub-point. The first point refers to the terminal that is the most matched with the sub-point and whose registered identity as the service provider of the sub-point is an insurance agent. The idle order points of each sequence refer to the set of order points that are not in service in the collection of order points of each sequence. The maximum order point overlap D2max refers to the maximum value of the order point overlap. The matching order point refers to the point that is the service provider of the sub-point in each sequence. The point series connection refers to the order of the service points of the sub-point obtained by sorting the first point and each matching order point in the order of service time.

[0098] Specifically, the association management module matches the sub-point with the most suitable service provider by calculating the overlap between the recommended portrait of the sub-point and the reverse portrait of each required point and sequence point, thereby ensuring service efficiency and user satisfaction, allowing users to obtain services that better meet their needs and preferences, thereby improving the overall user experience and service quality. Through automated association calculation and matching, human intervention is reduced, service efficiency and response speed are significantly improved, ensuring that users can obtain the required services in the shortest time. The association management module updates the recommended portrait and reverse portrait based on real-time data, making point matching more dynamic and flexible, and able to adapt to changes in user needs in a timely manner.

[0099] Specifically, the association management module compares the number of idle points n, and determines the optimization of the point association process according to the comparison result, wherein:

[0100] When n>1, the association management module determines not to optimize the point association process;

[0101] When n=1, the association management module determines not to optimize the point association process;

[0102] When n=0, the association management module determines to optimize the point association process, and uses the earliest released point as the matching point of the sub-point and associates it with the sub-point.

[0103] Specifically, the number of idle points refers to the number of required points that are in an idle state or the number of sequence points that are in an idle state in each sequence, the optimization of the point association process refers to whether the result of the point association performed by the association management module on the points that have passed the identity authentication is adjusted, the earliest released point refers to the sequence point or required point with the shortest time interval from being in an idle state, and the matching point refers to the sequence point or required point that matches the sub-point after optimizing the point association process.

[0104] Specifically, the association management module judges the optimization of the point association process of each sub-point and the required point and the sequential points of each order respectively, so as to dynamically adjust the point association process according to the number of idle points n, so as to make resource scheduling more intelligent. When n is 0, the earliest released point is automatically selected for association to ensure the most effective use of resources. The dynamic optimization mechanism enables users to receive timely services even when resources are tight, so as to improve user experience and satisfaction.

[0105] Specifically, the association management module compares the association discrimination Q with the preset association discrimination Q0, and makes an accuracy judgment on the optimization of the point association process according to the comparison result, wherein:

[0106] When Q≥Q0, the association management module determines that the optimization of the point association process is accurate, and outputs it as the accuracy judgment result;

[0107] When Q<Q0, the association management module determines that the optimization of the point association process is inaccurate, and outputs it as the accuracy judgment result.

[0108] Specifically, the association discrimination degree Q is set according to different subjects for accuracy judgment. When the accuracy of the optimization of the first point association process is judged, the association discrimination degree Q refers to the difference between the maximum first point overlap degree D1max and the sequential first point overlap degree D1a. The sequential first point overlap degree D1a refers to the maximum first point overlap degree value smaller than the maximum first point overlap degree D1max except the maximum first point overlap degree D1max. When judging the accuracy of the optimization of the association process of sequence points, the association discrimination Q refers to the difference between the maximum sequence point overlap D2max and the sequential sequence point overlap D2a, the sequential sequence point overlap D2a refers to the value of the maximum sequence point overlap that is smaller than the maximum sequence point overlap D2max except the maximum sequence point overlap D2max, and the preset association discrimination refers to the preset value of the association discrimination indicating that the optimization of the point association process is inaccurate, such as 0.25.

[0109] Specifically, the association management module adjusts the optimization process of the point association process according to the accuracy judgment result, wherein:

[0110] When the accuracy judgment result indicates that the optimization of the point association process is accurate, the association management module does not adjust the optimization process of the point association process;

[0111] When the accuracy judgment result is that the optimization situation of the point association process is inaccurate, the association management module adjusts the optimization process of the point association process, and when n>1, the association management module adjusts the judgment result of the optimization situation of not optimizing the point association process to optimizing the point association process, and performs association adjustment on the target point corresponding to the accuracy judgment result that the optimization situation of the point association process is inaccurate;

[0112] When the target point is adjusted for association, the association management module optimizes and iterates the preset statistical weight matrix of the point overlap of the target point until the association discrimination of the target point overlap is greater than or equal to the preset association discrimination, and records the number of optimization iterations Y, compares the number of optimization iterations Y with the preset number of optimization iterations Y0, and corrects the association adjustment according to the comparison result, wherein:

[0113] When Y≥Y0, the association management module corrects the association adjustment, increases the image parameters, and optimizes and iterates the preset statistical weight matrix of the point coincidence of the target points according to the increased image parameters until the association discrimination of the target point coincidence is greater than or equal to the preset association discrimination;

[0114] When Y<Y0, the association management module does not correct the association adjustment.

[0115] Specifically, the target point refers to the first point or each sequential point whose accuracy judgment result is that the optimization situation of the point association process is inaccurately corresponding, the point coincidence of the target point refers to the point coincidence of the first point or each sequential point and the sub-point whose accuracy judgment result is that the optimization situation of the point association process is inaccurately corresponding, and the optimization iteration refers to a method of adjusting the preset statistical weight matrix of the point coincidence of the target point. This embodiment does not limit the method of the optimization iteration, and those skilled in the art can freely set it according to actual conditions, and only needs to meet the need to adjust the preset statistical weight matrix of the point coincidence of the target point. For example, the point weight matrix of the target point can be adjusted by a random number generator. The preset statistical weight matrix of the coincidence degree of the target points is adjusted, the associated discrimination degree of the coincidence degree of the target points, the number of optimization iterations refers to the number of times the preset statistical weight matrix of the coincidence degree of the target points is adjusted, the preset number of optimization iterations refers to the preset value of the number of optimization iterations that indicates that the number of optimization iterations is large and the associated adjustment needs to be corrected, such as it can be set to 10 times, the portrait parameter refers to the data that reflects the service characteristics and quality of the target points, this embodiment does not limit the way of increasing the portrait parameters, and those skilled in the art can set them according to actual conditions, and only need to meet the needs of data screening on the service characteristics and quality of the target points, such as the portrait parameters can be increased through the Pearson correlation coefficient.

[0116] Specifically, the association management module compares the association discrimination to ensure that the association process is accurate only when Q≥Q0, effectively improving the accuracy of point association and avoiding the occurrence of erroneous association. When the association process is determined to be inaccurate, the optimization process is automatically adjusted to ensure that the optimization process of point association is improved, thereby improving the flexibility and adaptability of the system. Through the optimization iteration mechanism, the preset statistical weight matrix of point overlap is continuously adjusted until the preset association discrimination requirement is met, thereby ensuring the continuous optimization of the point association process.

[0117] Specifically, the execution management module manages the service execution of the first point and the sequential points according to the sub-point service process, wherein:

[0118] When the sub-point service process is associated, the execution management module notifies the head point of the sub-point to provide service;

[0119] When the sub-point service process is in the current order to be scheduled, the execution management module does not notify the sub-points in the current order to provide services;

[0120] When the sub-point service process is in progress for the current order, the execution management module notifies the sub-points of the current order to provide services;

[0121] When the sub-point service process is completed for the current order, the execution management module does not notify the sub-points of the current order to provide services.

[0122] Specifically, the sub-point service process refers to the service process of the sub-point at the current time, including association completed, current order to be scheduled, current order in progress and current order completed. The association completed refers to the time point after the point association is performed on the points that have passed the identity authentication. The current order to be scheduled refers to the time point when the current order point is in the process to be scheduled. The current order in progress refers to the time point when the current order point is in the process in progress. The current order completed refers to the time point when the current order point is in the completed process. This embodiment does not limit the method for obtaining the sub-point service process, and those skilled in the art can obtain the sub-point service process according to actual conditions. The situation can be freely set, and it is only necessary to meet the precise positioning requirement of the sub-point service process at the current time. For example, the option of obtaining the service process of the sub-point can be set to obtain the sub-point service process. This embodiment does not limit the way in which the execution management module notifies the first point of the sub-point and the sequence point of the current order. Technical personnel in this field can freely set it according to the actual situation, and it is only necessary to meet the obvious prompt requirement for the first point of the sub-point and the sequence point of the current order. For example, it can be set to push the service start reminder to the terminal of the first point of the above-mentioned sub-point and the sequence point of the current order through a push window.

[0123] Specifically, the execution management module compares the service point response time T with the preset service point response time T0, and optimizes the first point and the sequence point of each sub-point according to the comparison result, wherein:

[0124] When T≤T0, the execution management module determines that the first point and the sequential point of each sub-point respond in time, and does not optimize the first point and the sequential point of each sub-point;

[0125] When T>T0, the execution management module determines that the first point and the sequential point of each sub-point do not respond in time, and optimizes the first point and the sequential point of each sub-point as the response optimization point;

[0126] When optimizing the response optimization point, the response optimization point is sent to the association management module, the matching response optimization point output by the association management module is obtained, and the matching response optimization point is used as the first point and the sequential point of each of the sub-points;

[0127] The execution management module obtains the service point response time T1 of the matching response optimization point, and compares the service point response time T1 of the matching response optimization point with the preset service point response time T0, and judges the optimization effect of the matching response optimization point according to the comparison result, wherein:

[0128] When T1≤T0, the execution management module determines that the optimization effect of the matching response optimization point is good;

[0129] When T1>T0, the execution management module determines that the optimization effect of the matching response optimization point is poor.

[0130] Specifically, the service point response time is the time elapsed from the time point when the first point of the sub-point provides service to the time point when the first point of the sub-point confirms to provide service, and refers to the time elapsed from the time point of notifying the sequence point of the current order or notifying the sequence point of the current order to the time point when the sequence point of the current order confirms to provide service. The service point response time of the matching response optimization point refers to the time elapsed from the time point when the matching response optimization point of the sub-point provides service to the time point when the matching response optimization point of the sub-point confirms to provide service. The preset service point response time refers to a preset value indicating the time period during which the first point and sequence point of each sub-point do not respond in time, such as 1 hour. The response optimization point refers to the first point and sequence point of each sub-point to be re-associated by the association management module. The matching response optimization point refers to the first point and sequence point matched with each sub-point obtained by the association management module after analysis based on the response optimization point.

[0131] Specifically, when the execution management module determines that the optimization effect of the matching response optimization point is poor, the emergency point with the smallest release time is selected from the emergency point group as the emergency optimization point of the first point and the sequential point of each sub-point.

[0132] Specifically, the emergency point group refers to a collection of points that specifically serve sub-points when it is determined that the optimization effect of the matching response optimization point is poor, the emergency point with the shortest release time refers to the emergency point with the shortest time in the idle state, and the emergency optimization point refers to an emergency point that replaces the matching response optimization point with poor optimization effect as the service point of each sub-point.

[0133] Specifically, the execution management module locates the service process status of the sub-point and promptly notifies the primary point and the sequential point to provide services, effectively reducing waiting time and improving service efficiency. When the matching response optimization point is not effective, the system selects the emergency point with the shortest release time from the emergency point group as a replacement, realizing dynamic adjustment and optimization to ensure the continuity and stability of the service.

[0134] Specifically, the point group supervision module obtains the total number of emergency points k1, the number of times each emergency point is used p1, the number of response point groups k2 and the number of times each response point is used p2, calculates the utilization rate r1 of a single emergency point and the utilization rate r2 of a single response point, sets r1=p1 / H1, H1 is the total available time of each emergency point, sets r2=p2 / H2, H2 is the total available time of each response point, and the point group supervision module compares the utilization rate r1 of a single emergency point and the utilization rate r2 of a single response point during the peak period with the preset point utilization rate r0, and identifies the bottleneck point according to the comparison results, wherein:

[0135] When r1≥r0, the point group supervision module determines that the emergency point is a bottleneck point;

[0136] When r1<r0, the point group supervision module determines that the emergency point is not a bottleneck point;

[0137] When r2≥r0, the point group supervision module determines that the point is a bottleneck point;

[0138] When r2<r0, the point group supervision module determines that the point is not a bottleneck point;

[0139] The point group supervision module obtains the bottleneck emergency point quantity ka1 and the bottleneck response point quantity ka2, and calculates the resource utilization rate β, setting β=0.7×ka1 / k1+0.3×ka2 / k2. The point group supervision module compares the resource utilization rate β with the preset resource utilization rate β0, and judges the point group supervision situation according to the comparison result, where:

[0140] When β≤β0, the point group supervision module determines that the point group supervision is normal, and does not optimize the number of emergency point groups and the number of response point groups;

[0141] When β>β0, the point group supervision module determines that the point group supervision situation is abnormal, and optimizes the number of emergency point groups and the number of response point groups. The optimized number of emergency point groups is kb1, and kb1=α×k1 is set. The optimized number of response point groups is kb2, and kb2=α×k2 is set. 1<α≤1.2 is set, and α is the point group quantity optimization coefficient.

[0142] Specifically, the total number of emergency points refers to the total number of emergency points in the emergency point group, the number of times each emergency point is used refers to the cumulative number of services provided by a single emergency point in the emergency point group, the number of emergency point groups refers to the total number of emergency points in the emergency point group, the number of times each emergency point is used refers to the cumulative number of services provided by a single emergency point in the emergency point group, the total available time of each emergency point refers to the cumulative service time of a single emergency point in the emergency point group, and the total available time of each emergency point refers to the cumulative service time of a single emergency point in the emergency point group. The preset point utilization rate refers to the preset value of the point utilization rate indicating that the emergency point or the response point is a bottleneck point, and it can be set to 90%. The peak period refers to the time period with high frequency of use of the emergency point or the response point, such as 6 pm to 8 pm. The number of bottleneck emergency points refers to the total number of emergency points determined to be bottleneck points, and the number of bottleneck response points refers to the total number of response points determined to be bottleneck points. The preset resource utilization rate refers to the preset value of the resource utilization rate indicating that the point group supervision situation is abnormal, and it can be set to 80%.

[0143] Specifically, the point group supervision module can accurately identify bottleneck points during peak hours based on the comparison of utilization rates, promptly discover and resolve service bottlenecks, improve the service efficiency of the overall system, and optimize resources in abnormal situations to ensure efficient use of point group resources and improved service quality, reduce the occurrence of bottleneck points, optimize resource allocation, ensure the timeliness and efficiency of services, and thus significantly improve user satisfaction and experience.

[0144] Specifically, the information management module sends the sub-point service information to the first point and the sequential points, and shares the first point service information and the sequential point service information to the sub-point terminals.

[0145] Specifically, the sub-point service information refers to information indicating sub-point service requirements, the first-point service information refers to information on the first point generated in response to sub-point service requirements, and the sequential-point service information refers to information on sequential points generated in response to sub-point service requirements.

[0146] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. An insurance information management system based on a neural network model, characterized in that: include: The login management module is used to identify the type of the point, to authenticate the identity of the point after the type identification, to store the data of the point that has passed the authentication in slices, to adjust the type identification result, and to perform information isolation judgment on the data slice storage of the point; Sharded storage database is used to store the data generated by the points; The association management module is used to associate the points that have passed the identity authentication, obtain the first point and the sequential point of each sub-point, optimize the point association process, determine the accuracy of the optimization of the point association process, and adjust the optimization process of the point association process; An execution management module is used to manage the service execution of the first point and the sequential point, and to optimize the first point and the sequential point of each sub-point, and to optimize the first point and the sequential point of each sub-point according to the emergency point group; The point group supervision module is used to judge the point group supervision situation according to the number of emergency point optimizations and the number of response point groups, and optimize the number of emergency point groups and the number of response point groups according to the judgment results; Information management module, used to manage the information transmission between sub-points, primary points and sequential points; The login management module obtains the user ID, user registration identity and user selected identity, and performs type identification on the point, wherein: When the number of user registration identities corresponding to the user ID in the shard storage database is U=1, the login management module uses the user registration identity corresponding to the user ID as the point type; When the number of user registered identities corresponding to the user ID in the shard storage database is greater than 1, the login management module pushes an identity selection window to the point and uses the user selected identity as the point type; The login management module outputs the point type as a type identification result; The login management module obtains the input password, and compares the input password with the preset password corresponding to the user ID in the shard storage database, and performs identity authentication on the point after type recognition according to the comparison result, wherein: When the input password is consistent with the preset password, the login management module determines that the point after the type identification has passed the identity authentication, and marks it as a point that has passed the identity authentication; When the input password is inconsistent with the preset password, the login management module determines that the point after the type recognition has not passed the identity authentication; The login management module stores the generated data in the target shard of the shard storage database; The login management module compares the point switching type option with the original point type, and determines the adjustment of the type recognition result according to the comparison result, wherein: When the point switching type option is consistent with the original point type, the login management module determines not to adjust the type recognition result; When the point switching type option is inconsistent with the original point type, the login management module determines to adjust the type recognition result and adjusts the type recognition result to the point switching type option; After adjusting the type recognition result, the login management module shares information between the target slice of the origin type and the target slice of the adjusted type recognition result, and performs relationship analysis on the user to obtain the user ID that has a vested interest in the user, and uses it as the relationship ID. The target slice of the relationship ID is informationally isolated from the target slice of the user, and the login management module repeats information sharing and information isolation for the user.

2. The insurance information management system based on the neural network model according to claim 1 is characterized in that: When the association management module associates the points that have passed the identity authentication, it generates the recommended portrait of each sub-point in the point group, the reverse portrait of each corresponding point in the corresponding point group, and the reverse portrait of each sequence point in each sequence point group, calculates the first point overlap D1 of the recommended portrait of the sub-point and the reverse portrait of each corresponding point, and the sequence point overlap D2 of the recommended portrait of the sub-point and the reverse portrait of each sequence point, and the association management module associates the corresponding point with the maximum first point overlap D1max as the first point of the sub-point, associates the sequence point with the maximum sequence point overlap D2max among the free sequence points of each sequence as the matching sequence point of the sub-point in each sequence, and connects the matching sequence points of each sequence in series to obtain the sequence points of the sub-point; The association management module compares the number of idle points n, and determines the optimization of the point association process according to the comparison result, wherein: When n>1, the association management module determines not to optimize the point association process; When n=1, the association management module determines not to optimize the point association process; When n=0, the association management module determines to optimize the point association process, and uses the earliest released point as the matching point of the sub-point and associates it with the sub-point.

3. The insurance information management system based on the neural network model according to claim 2 is characterized in that: The association management module compares the association discrimination Q with the preset association discrimination Q0, and makes an accuracy judgment on the optimization of the point association process according to the comparison result, wherein: When Q≥Q0, the association management module determines that the optimization of the point association process is accurate, and outputs it as the accuracy judgment result; When Q<Q0, the association management module determines that the optimization of the point association process is inaccurate, and outputs it as the accuracy judgment result; The association management module adjusts the optimization process of the point association process according to the accuracy judgment result, wherein: When the accuracy judgment result indicates that the optimization of the point association process is accurate, the association management module does not adjust the optimization process of the point association process; When the accuracy judgment result is that the optimization situation of the point association process is inaccurate, the association management module adjusts the optimization process of the point association process, and when n>1, the association management module adjusts the judgment result of the optimization situation of not optimizing the point association process to optimizing the point association process, and performs association adjustment on the target point corresponding to the accuracy judgment result that the optimization situation of the point association process is inaccurate; When the target point is adjusted for association, the association management module optimizes and iterates the preset statistical weight matrix of the point overlap of the target point until the association discrimination of the target point overlap is greater than or equal to the preset association discrimination, and records the number of optimization iterations Y, compares the number of optimization iterations Y with the preset number of optimization iterations Y0, and corrects the association adjustment according to the comparison result, wherein: When Y≥Y0, the association management module corrects the association adjustment, increases the image parameters, and optimizes and iterates the preset statistical weight matrix of the point coincidence of the target points according to the increased image parameters until the association discrimination of the target point coincidence is greater than or equal to the preset association discrimination; When Y<Y0, the association management module does not correct the association adjustment.

4. The insurance information management system based on the neural network model according to claim 1, characterized in that: The execution management module manages the service execution of the first point and the sequential points according to the sub-point service process, wherein: When the sub-point service process is associated, the execution management module notifies the head point of the sub-point to provide service; When the sub-point service process is in the current order to be scheduled, the execution management module does not notify the sub-points in the current order to provide services; When the sub-point service process is in progress for the current order, the execution management module notifies the sub-points of the current order to provide services; When the sub-point service process is completed for the current order, the execution management module does not notify the sub-points of the current order to provide services; The execution management module compares the service point response time T with the preset service point response time T0, and optimizes and determines the first point and the sequence point of each sub-point according to the comparison result, wherein: When T≤T0, the execution management module determines that the first point and the sequential point of each sub-point respond in time, and does not optimize the first point and the sequential point of each sub-point; When T>T0, the execution management module determines that the first point and the sequential point of each sub-point do not respond in time, and optimizes the first point and the sequential point of each sub-point as the response optimization point; When optimizing the response optimization point, the response optimization point is sent to the association management module, the matching response optimization point output by the association management module is obtained, and the matching response optimization point is used as the first point and the sequential point of each of the sub-points; The execution management module obtains the service point response time T1 of the matching response optimization point, and compares the service point response time T1 of the matching response optimization point with the preset service point response time T0, and judges the optimization effect of the matching response optimization point according to the comparison result, wherein: When T1≤T0, the execution management module determines that the optimization effect of the matching response optimization point is good; When T1>T0, the execution management module determines that the optimization effect of the matching response optimization point is poor; When the execution management module determines that the optimization effect of the matching response optimization point is poor, the emergency point with the smallest release time is selected from the emergency point group as the emergency optimization point of the first point and the sequential point of each sub-point.

5. The insurance information management system based on the neural network model according to claim 4 is characterized in that: The point group supervision module obtains the total number of emergency points k1, the number of times each emergency point is used p1, the number of response point groups k2 and the number of times each response point is used p2, calculates the utilization rate r1 of a single emergency point and the utilization rate r2 of a single response point, sets r1=p1 / H1, H1 is the total available time of each emergency point, sets r2=p2 / H2, H2 is the total available time of each response point, and compares the utilization rate r1 of a single emergency point and the utilization rate r2 of a single response point during the peak period with the preset point utilization rate r0, and identifies the bottleneck point according to the comparison result, wherein: When r1≥r0, the point group supervision module determines that the emergency point is a bottleneck point; When r1<r0, the point group supervision module determines that the emergency point is not a bottleneck point; When r2≥r0, the point group supervision module determines that the point is a bottleneck point; When r2<r0, the point group supervision module determines that the point is not a bottleneck point; The point group supervision module obtains the bottleneck emergency point quantity ka1 and the bottleneck response point quantity ka2, and calculates the resource utilization rate β, setting β=0.7×ka1 / k1+0.3×ka2 / k2. The point group supervision module compares the resource utilization rate β with the preset resource utilization rate β0, and judges the point group supervision situation according to the comparison result, where: When β≤β0, the point group supervision module determines that the point group supervision is normal, and does not optimize the number of emergency point groups and the number of response point groups; When β>β0, the point group supervision module determines that the point group supervision situation is abnormal, and optimizes the number of emergency point groups and the number of response point groups. The optimized number of emergency point groups is kb1, and kb1=α×k1 is set. The optimized number of response point groups is kb2, and kb2=α×k2 is set. 1<α≤1.2 is set, and α is the point group number optimization coefficient; The information management module sends the sub-point service information to the first point and the sequential points, and shares the first point service information and the sequential point service information to the sub-point terminals.

Images