TWI868359B - Claims Decision-making System - Google Patents

Abstract

The present invention relates to a claim decision-making assistance system, including an input module, a claim decision module, a database service module, and a decision suggestion provision and tracking module. The input module is used to input case information. The claim decision module makes a graded suggestion based on a case feature point and an estimated case closing time for the input case information, and makes a loss determination and compensation amount decision for the case information with the graded suggestion. The database service module provides case history data to the claim decision module. The decision suggestion provision and tracking module publishes the loss determination and compensation amount decision to the system or user, and tracks the subsequent progress of the case.

Description

Claims decision support system

The present invention relates to a decision-making system, and more specifically, to a claims decision-making assistance system.

In the modern insurance industry, there is a trend of substantial growth in terms of case types, quantity, new employees, and insurance categories. Insurance claim adjusters often classify such diverse cases by department and make divisions of responsible units.

Traditionally, different cases are handled through "case classification" and "case damage assessment". By searching similar historical cases, the responsible personnel will classify and assess the damages of new cases and estimate the amount of compensation, and then submit them to the superior unit for approval before entering the compensation stage. However, the above procedures are likely to cause the following problems: 1. It takes time to search for historical cases; 2. If there is no previous case as a reference for a new case, it will take more time to estimate the amount of compensation; 3. New insurance personnel cannot classify and assess the case by themselves; 4. Insurance companies cannot analyze the causes of customer accidents and optimize the compensation process, so they cannot launch accurate insurance services for specific customers.

In addition, in the existing Republic of China New Patent Certificate No. M591674, a system for assisting underwriting of insurance claims is disclosed, which can first extract keyword strings from medical records, and after constructing the grammatical structure of medical records, respectively compare the medical terms and disease standard codes with the closest meanings in the reference tables of medical terms and disease classifications, and then determine whether the disease name corresponding to the disease standard code appears in the insurance terms. Subsequently, the system compares the keywords in the medical records with the disease standard codes. The keyword string is assigned a first identification mark (such as a highlight mark), and then a second identification mark corresponding to the clause matching result is assigned to an auxiliary judgment reference table generated by the system. Finally, an electronic medical record embedded with the first identification mark and an auxiliary judgment reference table embedded with the second identification mark are integrated into an intelligent medical record, thereby improving the review efficiency of underwriting or claims personnel, and at the same time effectively reducing the possibility of missed judgments and misjudgment of claims. However, in this patent, insurance companies cannot analyze the causes of customer accidents and optimize the claims process.

The existing Republic of China invention patent certificate No. I676954 discloses an automatic claim settlement system and method thereof, which includes a claim settlement server, and the claim settlement server includes a storage unit, a communication unit, and a processing unit. The processing unit transmits a data request to a data open platform server via the communication unit, so that when the data open platform server determines that it stores a release data of the day, it returns the release data of the day to the processing unit. After receiving the release data, the processing unit determines whether the subject information and at least one claim item of the insurance data meet a claim condition for each insurance data according to the release data, and for each insurance data determined to meet the claim condition, the processing unit performs a remittance transaction according to the insurance data and the insured's account information. In addition, the present invention also provides an automatic claim method. However, in this patent, if the new case does not have a previous case as a standard, it will take more time to estimate the claim amount.

In the existing Republic of China New Patent Certificate No. M585952, an insurance claim system using deep learning technology is disclosed, wherein the system is used to perform text recognition on the insurance claim document to generate a claim text file, then perform vocabulary recognition on the claim text file and generate a vocabulary recognition result, and generate a word vector based on the vocabulary recognition result. The word vector is generated by a convolutional neural network of deep learning to generate a high-dimensional feature matrix. Then, an insurance type code is received, and a deep learning classification operation is performed based on the high-dimensional feature matrix and the insurance type code of the policyholder's policy to generate a claim disposal code. However, in this patent, if a new case does not have a previous case as a reference, it will take more time to estimate the amount of compensation.

In the existing ROC invention patent application No. 202020791, a method and device for processing claims is disclosed. The method for processing claims includes: obtaining claim data in a claim application; obtaining claim rules corresponding to the policy information of the policy to be claimed based on the policy identification; and obtaining the credit information of the claim user based on the user identification; determining a number of rule features to be extracted based on the correspondence between the preset claim rules and the rule features, and extracting the rule features from the claim certificate; and extracting a number of preset credit features from the credit data; inputting the extracted features into a pre-trained claim review model, and determining whether the claim application is passed based on the output of the model. However, in this patent, if a new case does not have a previous case as a reference, it will take more time to estimate the amount of compensation.

In the existing Republic of China invention patent certificate No. I715886, a claim review device and a claim review method are disclosed. The claim review device includes an insurance database and a processor, wherein the processor is coupled to the insurance database. The insurance database is used to store historical claim data, and the historical claim data includes a claim type and a claim amount corresponding to the claim type. The processor is used to receive a claim request, read historical claim data from the insurance database, determine whether the claim request has a certain claim type, and if the processor determines that the claim request has a certain claim type, obtain the claim amount corresponding to the claim type. However, in this patent, if a new case does not have a previous case as a standard, it will take more time to estimate the claim amount.

In the existing Republic of China New Patent Certificate No. M560649, an image recognition-assisted claims system is disclosed, which includes an imaging module, an identification unit, a processing unit and a storage unit. The imaging module captures a claim document to generate a claim document image file corresponding to the claim document; the identification unit identifies a pathology name in the claim document image file; the processing unit is electrically connected to the imaging module and the identification unit, and the processing unit obtains a corresponding International Classification of Diseases code according to the pathology name; the storage unit is electrically connected to the processing unit to store the International Classification of Diseases code, and automatically identifies the pathology name and the International Classification of Diseases code, which can greatly save the working time of the claims personnel. However, under this patent, insurance companies cannot analyze the causes of customer accidents and optimize the claims process, and therefore cannot launch insurance services that are accurate for specific customers.

The existing Republic of China invention patent certificate No. I684157 discloses an intelligent claims system based on a mobile vehicle, which includes a mobile terminal device, a cloud resolution server and an insurance claims server, wherein the mobile terminal device includes an image capture module, a seal processing module, a character recognition module and a mobile communication module; the cloud resolution server includes a cloud communication unit, a cloud storage unit and a cloud processing unit; and the insurance claims server includes a claims communication unit, a claims storage unit and a claims processing unit for generating the claims results and the claims amounts of the insurance policies. However, in this patent, if a new case does not have a previous case as a reference, it will take more time to estimate the amount of compensation.

In the existing Republic of China New Patent Certificate No. M593619, a real-time claims system is disclosed, which includes a server host and a mobile device. The mobile device generates a plurality of application data through an image capture module. The server host includes an image recognition module, an intelligent code selection module, a risk determination module, and a claims verification module. When the server receives the application data, the image recognition module performs image recognition on each application data to obtain an application content corresponding to each application data. The intelligent code selection module selects multiple corresponding system codes according to each application content. The risk judgment module judges a risk level of the corresponding claim case according to each application content, the system codes, and multiple judgment factors. The claim assessment module determines whether to generate a remittance instruction based on the risk level. However, in this patent, new insurance personnel cannot classify and determine the loss of the case by themselves.

In the existing Republic of China invention patent certificate No. I710997, a method and system for automatic car insurance compensation is disclosed. The method for automatic car insurance compensation includes guiding the user to take a vehicle image and obtain the vehicle image; performing image recognition on the obtained vehicle image to identify the vehicle damage; generating a repair list based on the vehicle damage; reviewing the user's credit when the user accepts the repair list; and automatically compensating the user when the user's credit is high enough. However, in this patent, the insurance company cannot analyze the cause of the customer's accident and optimize the compensation process, so it cannot launch insurance services that are accurate for specific customers.

In the existing ROC invention patent application No. 202025058, an insurance claim application system is disclosed, including an input module, a character recognition module, a text parsing module, and a policy analysis module. The input module is suitable for inputting a document image. The character recognition module receives the document image and converts it into a text data. The text parsing module is connected to the character recognition module, receives the text data, and extracts a plurality of necessary information from the text data. The policy analysis module is connected to the text parsing module, receives and analyzes the necessary information, and generates a claim information. However, in this patent, if a new case does not have a previous case as a standard, it will take more time to estimate the claim amount.

In view of this, how to provide a claims decision-making assistance system that can find previous cases in a more efficient way, assist insurance personnel in classifying and assessing cases, and further analyze the causes of customer accidents and optimize the claims process, so as to launch insurance services accurately targeting specific customers in the future, is an issue that the industry urgently needs to solve.

In order to solve the above-mentioned deficiencies of the prior art, the main purpose of the present invention is to provide a claim decision-making assistance system, which includes an input module, a claim decision module, a database service module and a decision suggestion provision and tracking module; the input module is used to input case information; the claim decision module has a case classification module and a case damage assessment module, and the case classification module has a case feature point that includes but is not limited to historical case data, claim experience, claim amount, experts, etc. The case damage assessment module completes a prediction model and establishes at least one case feature label through the content of the case historical data, and continuously tracks the subsequent execution progress of the case. Through the case historical data and claims experience, the case features are found, the case closing time is estimated, and then the case classification recommendations and decisions are made. The case classification module will make a classification recommendation based on the input case information and the case features and estimated case closing time. The case damage assessment module The case information with the classification recommendation is used to make a decision on the amount of damage and compensation; the claim decision-making system establishes and trains a judgment logic model based on the judgment method of the case history record through methods including but not limited to machine learning and neural network learning, as one of the basis for algorithm decision-making; the database service module provides the case history data to the claim decision-making module, and serves as the algorithm calculation model source of the case classification module and the case damage assessment module; the decision recommendation is provided and tracked The module publishes the damage assessment and compensation amount decision to the system or users, and tracks the subsequent progress of the case; the case information possessed by the claim decision-making system includes but is not limited to accident data, object data and insurance type data; the accident data possessed by the claim decision-making system includes but is not limited to accident personnel information, case evidence, accident content, and police handling records; the insurance type data possessed by the claim decision-making system includes but is not limited to insurance type, service requirements, business content and attributes.

To achieve the above purpose, the decision-making suggestion provision and tracking module of the claims decision-making auxiliary system of the present invention includes a suggestion release module and a case tracking module. The suggestion release module provides the claim method and process to a case person in charge for the completed damage assessment and compensation amount decision, and the case tracking module tracks the subsequent claim process for the case that has completed the decision suggestion.

To achieve the above purpose, the object data of the claims decision-making assistance system of this invention includes but is not limited to name, ID number and information, telephone number, age, birthday, and address.

To achieve the above purpose, the case history data of the claims decision-making assistance system of this invention includes but is not limited to underwriting history, expert experience, judgments, claims process, claims case data, data of all persons related to the claims case, case processing schedule and case closing time records, decision-making key points and evidence, claims decision and loss assessment records of compensation amount and method.

Through this system, when the claimant is not sure how to make a decision or needs to infer the method and amount of damages based on historical cases, there is a systematic solution to provide the claimant with a quick decision and enter the claim process. This system can calculate all historical cases and infer the claim decision of new cases, and strengthen the new cases when there are no historical cases for reference. The system can also calculate the appropriate claim case classification and damages, achieving the goal of fully automated case classification and damages.

100: Claims decision-making support system

200: Input module

300: Claims decision module

310:Case classification module

320: Case damage assessment module

400: Database service module

500: Decision suggestion provision and tracking module

510: Suggest to publish module

520:Case tracking module

600:Case information

Figure 1 is a module structure diagram of the claims decision-making assistance system of the present invention; and Figure 2 is a module structure diagram of the decision-making suggestion providing and tracking module of the claims decision-making assistance system of the present invention.

The present invention is described in detail as follows with the accompanying drawings and in the form of an embodiment. The drawings used in the text are only for illustration and auxiliary description, and may not be the actual proportion and precise configuration of the present invention after implementation. Therefore, the proportion and configuration of the attached drawings should not be used to limit the patent scope of the present invention in actual implementation.

As shown in FIG1 , a claim decision-making auxiliary system 100 of the present invention includes an input module 200, a claim decision module 300, a database service module 400, and a decision suggestion providing and tracking module 500.

The input module 200 is used to input case information 600. The claim decision module 300 has a case classification module 310 and a case damage assessment module 320. The case classification module 310 makes a classification suggestion for the input case information 600 according to a case feature point and an estimated case closing time, and the case damage assessment module 320 makes a damage and compensation amount decision for the case information with the classification suggestion. The database service module 400 provides a case history data to the claim decision module 300 and serves as the algorithm calculation model source of the case classification module 310 and the case damage assessment module 320. The decision suggestion providing and tracking module 500 publishes the damage assessment and compensation amount decision to the system or users, and tracks the subsequent progress of the case; the case information 600 possessed by the claim decision-making auxiliary system 100 includes but is not limited to accident data, object data and insurance type data; the accident data possessed by the claim decision-making auxiliary system 100 includes but is not limited to accident personnel information, case evidence, accident content, police handling Records; the insurance data of the claims decision-making support system 100 includes but is not limited to insurance types, service requirements, business content and attributes; the case history data includes but is not limited to underwriting history, expert experience, judgments, claims process, claims case data, data of all persons related to the claims case, case processing schedule and case closing hours, decision-making key points and evidence, claims decision and loss assessment records of the amount and method of compensation.

Please refer to FIG. 2 . The decision suggestion providing and tracking module 500 of the claim decision-making auxiliary system 100 of the present invention includes a suggestion publishing module 510 and a case tracking module 520 . The suggestion publishing module 510 provides a claim method and process to a case person in charge for the completed damage assessment and compensation amount decision, and the case tracking module 520 tracks the subsequent claim process for the case for which the decision suggestion is completed.

It should be noted that the case information of the claims decision-making assistance system 100 of the present invention includes but is not limited to accident data, object data and insurance type data.

The aforementioned accident data includes but is not limited to accident personnel information, case evidence, accident content, police handling records, etc. The aforementioned object data includes but is not limited to name, ID number and information, telephone number, age, birthday, address, etc. The aforementioned insurance type data includes but is not limited to insurance type, service requirements, business content and attributes, etc.

Furthermore, the input module 200 of the claims decision-making assistance system 100 of the present invention may include a system terminal, and the system terminal includes but is not limited to devices or interfaces such as smart phones, tablets, computers, and browsers.

The case feature points of the claim decision-making assistance system 100 of the present invention are defined by including but not limited to historical case data, claim experience, claim amount, expert experience, judgment cases and decision records. In addition, the case damage assessment module 320 of the claim decision-making assistance system 100 of the present invention completes a prediction model through the content of the case historical data and establishes at least one case feature label (such as the claim amount is 6,000, and the expert experience is that the probability of claim is high), etc. All cases can have multiple feature labels. The case information of the claim decision-making assistance system includes but is not limited to accident data, object data and insurance type data; the accident data of the claim decision-making assistance system includes But not limited to the accident personnel information, case evidence, accident content, police handling records; the insurance data of the claim decision-making system includes but is not limited to insurance types, service requirements, business content and attributes; the case history data includes but is not limited to underwriting history, expert experience, judgment, claim process, claim case data, the data of all persons related to the claim case, case processing schedule and case closing time records, decision-making key points and evidence, claim decision and loss assessment records of compensation amount and method.

The aforementioned case history data includes but is not limited to underwriting history, expert experience, judgments, claims process, claims case data, data of all persons involved in the claims case, case processing schedule and case closing time records, decision-making key points and evidence, claims decision and loss assessment records of compensation amount and method, etc.

In summary, through the setting of this claims decision-making assistance system 100, when the claims personnel are not sure how to make a decision or need to infer the method and amount of damages based on historical cases, there will be a systematic solution to provide the claims personnel with a quick decision and enter the claims process. This system can strengthen the new cases when there are no historical cases for reference, and the system can also calculate the appropriate claims case classification and damages, so as to achieve the goal of fully automated case classification and damages.

In addition to the above functions, the claims decision-making assistance system 100 of the present invention can also have the following extended applications:

1. Notification of decision suggestions for new claims items: The system can update the prediction model data for the newly provided claims services and their corresponding decision methods, and notify users through methods including but not limited to text messages, voice, e-mail, etc., to convey the updated information of this claims service and the suggested decision-making behavior for related cases.

2. Real-time case progress and real-time contact and response system for each link: The system can track and query the case tracking codes of cases accepted by this system, including but not limited to the tracking codes of external cases. At the same time, it can also calculate appropriate subsequent claims decisions for cases that have not yet completed claims, and simultaneously update the case claims steps.

3. Common Q&A and systematic solution suggestions for claims clients: The system records all questions and answers during the claims decision-making process. The system will provide systematic solution suggestions for issues that clients are likely to mention, and provide clients with a simple Q&A table of common questions to reduce the repetitive responses and work of claims personnel.

4. Logic model building for claim settlement system: The system builds a logic model based on the judgment method of historical case records, including but not limited to machine learning, neural network learning, etc., and conducts training as one of the decision-making bases for the algorithm.

5. Post-insurance customer accident timeline model, with client recommendation system: The system establishes parameters for the period from when the client purchases insurance until the actual accident. Parameters include but are not limited to the insurance date, time, time from insurance purchase to accident occurrence, accident type, etc. The system can calculate key points after insurance purchase through the model, and can remind the insured customer of points to note and recommend things to avoid through notification methods including but not limited to e-mail, text messages, physical letters, etc.

The above-mentioned embodiments are only for illustrating the technical ideas and features of the present invention. Their purpose is to enable people familiar with this technology to understand the content of the present invention and implement it accordingly. They cannot be used to limit the patent scope of the present invention. In other words, any equivalent changes or modifications made according to the spirit disclosed by the present invention should still be covered by the patent scope of the present invention.

100: Claims decision-making support system

200: Input module

300: Claims decision module

310:Case classification module

320: Case damage assessment module

400: Database service module

500: Decision suggestion provision and tracking module

600:Case information

Claims (2)

A claims decision-making system includes: an input module, a claims decision module, a database service module, and a decision suggestion provision and tracking module; the input module is used to input case information; the claims decision module has a case classification module and a case loss assessment module, the case classification module has a case feature point defined by including but not limited to historical case data, claims experience, claims amount, expert experience, judgment cases and decision records, the case history data includes but not limited to underwriting history, expert experience, and judgment records. The database service module provides the historical data of the case to the claim decision module and serves as the algorithm calculation model source of the case classification module and the case damage assessment module. The characteristic of the module is that the case damage assessment module completes a prediction model and establishes at least one case feature label through the content of the case historical data, and continuously tracks the case. The subsequent execution progress of the case is to find out the characteristic points of the case and estimate the time required to close the case through the historical data of the case and the claims settlement experience, and then make case classification suggestions and decisions. The case classification module will make a classification suggestion based on the input case information according to the characteristic points of the case and the estimated time required to close the case. The case damage assessment module will make a certain loss and compensation amount decision based on the case information with the classification suggestion. The claims settlement auxiliary decision-making system is established based on the judgment method of the historical records of the case through methods including but not limited to machine learning, neural network learning, etc. Determine the logic model and train it as one of the decision bases for the algorithm; and the decision suggestion provision and tracking module publishes the damage assessment and compensation amount decision to the system or users, and tracks the subsequent progress of the case; the case information possessed by the claim settlement auxiliary decision system includes but is not limited to an accident data, an object data and an insurance type data; the accident data includes but is not limited to accident personnel information, case evidence, accident content, police handling records; the insurance type data includes but is not limited to insurance type, service requirements, business content and attributes. The claim decision-making assistance system as described in claim item 1, wherein the decision suggestion provision and tracking module includes a suggestion release module and a case tracking module, wherein the suggestion release module provides a claim method and process to a case person in charge for the completed damage assessment and compensation amount decision, and the case tracking module tracks the subsequent claim process for the case for which the decision suggestion has been completed.