CN115439265A - Intelligent insurance industry compensation abnormal transaction risk control system - Google Patents
Intelligent insurance industry compensation abnormal transaction risk control system Download PDFInfo
- Publication number
- CN115439265A CN115439265A CN202211173819.9A CN202211173819A CN115439265A CN 115439265 A CN115439265 A CN 115439265A CN 202211173819 A CN202211173819 A CN 202211173819A CN 115439265 A CN115439265 A CN 115439265A
- Authority
- CN
- China
- Prior art keywords
- transaction
- wind control
- account
- data
- abnormal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/08—Insurance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Finance (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses an insurance industry intelligent claim abnormal transaction risk control system, which comprises a transaction sample acquisition subsystem, a transaction processing subsystem and a transaction processing subsystem, wherein the transaction sample acquisition subsystem is used for acquiring transaction sample data in an insurance claim settlement process, and the transaction sample data comprises sample set data, account basic information data, a fund transaction map and transaction link data; the wind control subsystem is used for carrying out risk judgment and wind control grade judgment according to the transaction sample data; the wind control subsystem comprises a first wind control submodule for processing and wind control judgment according to the sample set data, a second wind control submodule for processing and wind control judgment according to the account basic information data, a third wind control submodule for processing and wind control judgment according to the fund transaction map, and a fourth wind control submodule for processing and wind control judgment according to the transaction work order data. According to the invention, the wind control subsystem carries out risk judgment and wind control grade judgment according to the multidimensional data, so that the accuracy of wind control detection of abnormal transactions is improved.
Description
Technical Field
The invention relates to the field of insurance claim payment, in particular to an intelligent claim payment abnormal transaction risk control system in the insurance industry.
Background
With the continuous development of socio-economy, people are more aware of the importance of insurance. The applicant pays an insurance premium to the insurer based on the contractual agreement, and the insurer undertakes compensation for the insurance funds for losses due to the potential risk of the contractual agreement. Therefore, the insurer is extremely important for risk assessment of the insurance business. When insurance claim risk assessment is performed in the prior art, an operator usually performs manual assessment aiming at a specific scene according to past experience of the operator, the individual judgment of the operator is mainly relied on, the subjectivity is strong, and the final assessment result is often low in accuracy. With increasingly fierce competition in insurance industry and increasingly strong customer service awareness, the traditional claim settlement service mainly has manual experience for risk management and control, the operation efficiency is too low, the leakage rate of risk cases is high, and the requirements of customers and insurance companies cannot be met, so that the demands of extremely-induced, rapid and accurate claim settlement and differentiated service are more and more urgent.
With the development of machine learning, but the current machine learning can only calculate the corresponding risk coefficient in advance by using the offline historical data of the same type of service according to the difference of services, and then performs data exchange with the database through an application program interface, and the accuracy and timeliness of the risk data cannot be guaranteed by the method.
Disclosure of Invention
The invention aims to provide an intelligent insurance industry abnormal transaction risk control system to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme: an insurance industry intelligent claim abnormal transaction risk control system, comprising: the system comprises a transaction sample acquisition subsystem, a transaction analysis subsystem and a transaction analysis subsystem, wherein the transaction sample acquisition subsystem is used for acquiring transaction sample data in the insurance claim settlement process, and the transaction sample data comprises sample set data, account basic information data, a fund transaction map and transaction link data; and the wind control subsystem is used for carrying out risk judgment and wind control grade judgment according to the transaction sample data.
Preferably, the wind control subsystem comprises a first wind control submodule for processing and wind control judgment according to sample set data, a second wind control submodule for processing and wind control judgment according to account basic information data, a third wind control submodule for processing and wind control judgment according to a fund transaction map, and a fourth wind control submodule for processing and wind control judgment according to transaction work order data,
the transaction sample acquisition subsystem is internally preset with a sample abnormal transaction identification model, and the sample abnormal transaction identification model performs invalid data removing operation on the historical abnormal transaction data and the transaction work order data according to the preset data integrity and obtains valid sample data; performing characteristic screening operation on the effective sample data to obtain training sample data; the second wind control sub-module can judge whether the current transaction conforms to the user portrait of the target user according to the target user label in the account basic information data; if the current transaction does not conform to the user portrait of the target user, performing exception handling on the target transaction; when the current transaction passes the transaction parameter verification, acquiring target account information to which the current transaction belongs, and determining target consumption information corresponding to a target user according to a target user tag in the target account information; comparing the current transaction information of the current transaction with the target consumption information to judge whether the current transaction conforms to the user portrait of the target user;
the third wind control submodule is used for acquiring node characterization data in a fund transaction map, the fund transaction map consists of nodes and edges, the node characterization and account data corresponding to the nodes, the edges between the two nodes characterize the transaction flow direction between the two nodes, and the third wind control submodule generates a node feature set based on the fund transaction map, wherein the node feature set consists of at least one node feature vector, and each node feature vector corresponds to one node in the fund transaction map; inputting the feature vector of each node into a pre-trained graph model by a third wind control sub-module to obtain a prediction result corresponding to each node, wherein the prediction result is used for representing whether an account corresponding to each node is a fund transaction abnormal account or not, and the graph model is obtained based on graph training different from a fund transaction graph; the fourth wind control submodule comprises a construction unit, an aggregation unit and an abnormality identification unit, wherein the construction unit is used for constructing a transaction network graph corresponding to the transaction account to be detected by taking the transaction account to be detected as a central node and taking the transaction account A in user equipment and the transaction account B in merchant equipment as outer nodes of the central node; the aggregation module is used for carrying out feature aggregation processing on the node features of all nodes in the transaction network graph and obtaining aggregation features corresponding to the transaction account to be detected; and the abnormality identification module performs abnormality identification on the transaction account to be detected based on the aggregation characteristics so as to determine whether the transaction account to be detected is an abnormal transaction account.
Preferably, the first wind control submodule acquires the training sample data, constructs an abnormal transaction identification model based on logistic regression, inputs the account basic information data into the abnormal transaction identification model to obtain an abnormal transaction identification result, and outputs the first abnormal transaction identification result.
Preferably, the fourth wind control sub-module further determines a plurality of transaction links associated with the transaction account to be detected when the transaction account to be detected is determined to be an abnormal transaction account, and performs link feature extraction on account features of the transaction accounts in the transaction links respectively, and determines that the link features are abnormal link features or normal link features.
Preferably, the wind control subsystem further comprises a wind control evaluation submodule, the wind control evaluation submodule performs index evaluation according to the first abnormal transaction identification result, the abnormal transaction account judgment result and the user portrait judgment result, and generates corresponding alarm prompt information when the evaluation index is greater than or equal to a preset alarm threshold value.
Compared with the prior art, the invention has the beneficial effects that:
according to the method, based on the acquired sample set data, the account basic information data, the fund transaction map, the transaction link data and the like, the wind control subsystem carries out risk judgment and wind control grade judgment according to the multi-dimensional data, so that the accuracy of abnormal transaction wind control detection is improved.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a schematic processing flow diagram of a first wind control submodule according to an embodiment of the present invention;
fig. 3 is a schematic processing flow diagram of a second wind control submodule according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-3, the present invention provides a technical solution: an insurance industry intelligent claim payment abnormal transaction risk control system comprises a transaction sample acquisition subsystem, a transaction processing subsystem and a transaction processing subsystem, wherein the transaction sample acquisition subsystem is used for acquiring transaction sample data in an insurance claim settlement process, and the transaction sample data comprises sample set data, account basic information data, a fund transaction map and transaction link data; and the wind control subsystem is used for carrying out risk judgment and wind control grade judgment according to the transaction sample data.
In the embodiment, the wind control subsystem comprises a first wind control submodule for processing and wind control judgment according to the sample set data, a second wind control submodule for processing and wind control judgment according to the account basic information data, a third wind control submodule for processing and wind control judgment according to the fund transaction map, and a fourth wind control submodule for processing and wind control judgment according to the transaction work order data,
the transaction sample acquisition subsystem is internally preset with a sample abnormal transaction identification model, and the sample abnormal transaction identification model performs invalid data removing operation on the historical abnormal transaction data and the transaction work order data according to the preset data integrity and obtains valid sample data; performing characteristic screening operation on the effective sample data to obtain training sample data; the second wind control sub-module can judge whether the current transaction conforms to the user portrait of the target user according to the target user label in the account basic information data; if the current transaction does not conform to the user portrait of the target user, performing exception processing on the target transaction; when the current transaction passes the transaction parameter verification, acquiring target account information to which the current transaction belongs, and determining target consumption information corresponding to a target user according to a target user tag in the target account information; comparing the current transaction information of the current transaction with the target consumption information to judge whether the current transaction conforms to the user portrait of the target user; the third wind control submodule is used for acquiring node characterization data in a fund transaction map, the fund transaction map consists of nodes and edges, the node characterization and account data corresponding to the nodes, the edges between the two nodes characterize the transaction flow direction between the two nodes, and the third wind control submodule generates a node feature set based on the fund transaction map, wherein the node feature set consists of at least one node feature vector, and each node feature vector corresponds to one node in the fund transaction map; inputting the feature vector of each node into a pre-trained graph model by a third wind control sub-module to obtain a prediction result corresponding to each node, wherein the prediction result is used for representing whether an account corresponding to each node is a fund transaction abnormal account or not, and the graph model is obtained based on graph training different from a fund transaction graph; the fourth wind control submodule comprises a construction unit, an aggregation unit and an abnormality recognition unit, wherein the construction unit is used for constructing a transaction network graph corresponding to the transaction account to be detected by taking the transaction account to be detected as a central node and taking the transaction account A in the user equipment and the transaction account B in the merchant equipment as outer nodes of the central node; the aggregation module is used for carrying out feature aggregation processing on the node features of all nodes in the transaction network graph and obtaining aggregation features corresponding to the transaction account to be detected; and the abnormality identification module performs abnormality identification on the transaction account to be detected based on the aggregation characteristics so as to determine whether the transaction account to be detected is an abnormal transaction account.
In this embodiment, the first wind control submodule acquires the training sample data, constructs an abnormal transaction identification model based on logistic regression, obtains an abnormal transaction identification result by inputting the account basic information data into the abnormal transaction identification model, and outputs the first abnormal transaction identification result.
In this embodiment, when determining that the transaction account to be detected is an abnormal transaction account, the fourth wind control sub-module further determines a plurality of transaction links associated with the transaction account to be detected, and performs link feature extraction on account features of the transaction accounts in the transaction links respectively, and determines that the link features are abnormal link features or normal link features.
In this embodiment, the wind control subsystem further includes a wind control evaluation sub-module, and the wind control evaluation sub-module performs index evaluation according to the first abnormal transaction identification result, the abnormal transaction account determination result, and the user portrait determination result, and generates corresponding alarm prompt information when the evaluation index is greater than or equal to a preset alarm threshold value.
In this embodiment, in order to process a large amount of transaction data in real time, an open source stream processing platform Kafka cluster is used as a processing architecture. The Kafka cluster is originally developed by Linkedin company, is a distributed, partition-supporting, multi-copy and high-throughput distributed publishing and subscribing message system, is a distributed message system based on the coordination of a distributed application program coordination service zookeeper, and has the greatest characteristic that a large amount of data can be processed in real time to meet various demand scenarios. The method is applied to a common server, can process hundreds of thousands of messages per second, and can be used for collecting and sending a large amount of event and log data with low time delay.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (5)
1. An insurance industry intelligent claim abnormal transaction risk control system, comprising:
the system comprises a transaction sample acquisition subsystem, a transaction analysis subsystem and a transaction analysis subsystem, wherein the transaction sample acquisition subsystem is used for acquiring transaction sample data in the insurance claim settlement process, and the transaction sample data comprises sample set data, account basic information data, a fund transaction map and transaction link data;
and the wind control subsystem is used for carrying out risk judgment and wind control grade judgment according to the transaction sample data.
2. The insurance industry intelligent claim abnormal transaction risk control system of claim 1, wherein: the wind control subsystem comprises a first wind control submodule for processing and wind control judgment according to sample set data, a second wind control submodule for processing and wind control judgment according to account basic information data, a third wind control submodule for processing and wind control judgment according to a fund transaction map, and a fourth wind control submodule for processing and wind control judgment according to transaction work order data,
the transaction sample acquisition subsystem is internally provided with a sample abnormal transaction identification model in advance, and the sample abnormal transaction identification model carries out invalid data elimination operation on the historical abnormal transaction data and the transaction work order data according to the preset data integrity and obtains valid sample data; performing characteristic screening operation on the effective sample data to obtain training sample data;
the second wind control sub-module can judge whether the current transaction conforms to the user portrait of the target user according to the target user label in the account basic information data; if the current transaction does not conform to the user portrait of the target user, performing exception processing on the target transaction; when the current transaction passes the transaction parameter verification, acquiring target account information to which the current transaction belongs, and determining target consumption information corresponding to a target user according to a target user tag in the target account information; comparing the current transaction information of the current transaction with the target consumption information to judge whether the current transaction conforms to the user portrait of the target user;
the third wind control submodule is used for acquiring node characterization data in the fund transaction map, the fund transaction map consists of nodes and edges, the node characterization and account data corresponding to the nodes, the edge between the two nodes characterizes the transaction flow direction between the two nodes, and the third wind control submodule generates a node feature set based on the fund transaction map, wherein the node feature set consists of at least one node feature vector, and each node feature vector corresponds to one node in the fund transaction map; inputting the feature vector of each node into a pre-trained graph model by a third wind control sub-module to obtain a prediction result corresponding to each node, wherein the prediction result is used for representing whether an account corresponding to each node is a fund transaction abnormal account or not, and the graph model is obtained based on graph training different from a fund transaction graph;
the fourth wind control submodule comprises a construction unit, an aggregation unit and an abnormality identification unit, wherein the construction unit is used for constructing a transaction network graph corresponding to the transaction account to be detected by taking the transaction account to be detected as a central node and taking the transaction account A in user equipment and the transaction account B in merchant equipment as outer nodes of the central node; the aggregation module is used for carrying out feature aggregation processing on the node features of all nodes in the transaction network graph and obtaining aggregation features corresponding to the transaction account to be detected; and the abnormality identification module performs abnormality identification on the transaction account to be detected based on the aggregation characteristics so as to determine whether the transaction account to be detected is an abnormal transaction account.
3. The insurance industry intelligent claim abnormal transaction risk control system of claim 2, wherein: the first wind control submodule acquires the training sample data, constructs an abnormal transaction identification model based on logistic regression, inputs the account basic information data into the abnormal transaction identification model to obtain an abnormal transaction identification result, and outputs the first abnormal transaction identification result.
4. The insurance industry intelligent claim abnormal transaction risk control system of claim 2, wherein: and the fourth wind control sub-module also determines a plurality of transaction links associated with the transaction account to be detected when the transaction account to be detected is determined to be an abnormal transaction account, respectively extracts the link characteristics of the account characteristics of the transaction account in each transaction link, and determines the link characteristics to be the abnormal link characteristics or the normal link characteristics.
5. The insurance industry intelligent claim abnormal transaction risk control system of claim 1, wherein: the wind control subsystem further comprises a wind control evaluation submodule, the wind control evaluation submodule carries out index evaluation according to the first abnormal transaction identification result, the abnormal transaction account judgment result and the user portrait judgment result, and when the evaluation index is larger than or equal to a preset alarm threshold value, corresponding alarm prompt information is generated.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211173819.9A CN115439265A (en) | 2022-09-26 | 2022-09-26 | Intelligent insurance industry compensation abnormal transaction risk control system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211173819.9A CN115439265A (en) | 2022-09-26 | 2022-09-26 | Intelligent insurance industry compensation abnormal transaction risk control system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115439265A true CN115439265A (en) | 2022-12-06 |
Family
ID=84248663
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211173819.9A Pending CN115439265A (en) | 2022-09-26 | 2022-09-26 | Intelligent insurance industry compensation abnormal transaction risk control system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115439265A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117171141A (en) * | 2023-11-01 | 2023-12-05 | 广州中长康达信息技术有限公司 | Data model modeling method based on relational graph |
-
2022
- 2022-09-26 CN CN202211173819.9A patent/CN115439265A/en active Pending
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117171141A (en) * | 2023-11-01 | 2023-12-05 | 广州中长康达信息技术有限公司 | Data model modeling method based on relational graph |
CN117171141B (en) * | 2023-11-01 | 2024-02-20 | 广州中长康达信息技术有限公司 | Data model modeling method based on relational graph |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021232588A1 (en) | Food safety risk assessment method, apparatus, device, and storage medium | |
CN108228706A (en) | For identifying the method and apparatus of abnormal transaction corporations | |
TWI690865B (en) | Transaction volume prediction method and device | |
CN103123712A (en) | Method and system for monitoring network behavior data | |
CN111476660A (en) | Intelligent wind control system and method based on data analysis | |
CN111008896A (en) | Financial risk early warning method and device, electronic equipment and storage medium | |
CN104573996A (en) | Automated quotation system for casting enterprises | |
CN112785427B (en) | Enterprise credit analysis system based on power data | |
CN113076353A (en) | Electric power accounting abnormity analysis method and system | |
CN115439265A (en) | Intelligent insurance industry compensation abnormal transaction risk control system | |
CN115018513A (en) | Data inspection method, device, equipment and storage medium | |
CN111695979A (en) | Method, device and equipment for analyzing relation between raw material and finished product | |
CN116151840B (en) | User service data intelligent management system and method based on big data | |
CN110910232A (en) | Financial reconciliation system and method | |
CN113506173A (en) | Credit risk assessment method and related equipment thereof | |
KR101666824B1 (en) | Chain bankruptcy risk management system and bankruptcy risk management method using the same | |
Khouilid et al. | The impact of Non-Tariff Measures (NTMs) on Moroccan foreign trade: Comparison between developed and developing countries | |
CN108961037A (en) | A kind of loan for vehicle air control method and device based on to vehicle service condition assessment algorithm | |
CN117455681A (en) | Service risk prediction method and device | |
CN113449923A (en) | Multi-model object market quotation prediction method and device | |
Abdullah et al. | Malaysian financial stress index and assessing its impacts on the economy | |
CN112199360A (en) | Data processing method, device, equipment and medium | |
CN113065736A (en) | Risk monitoring method, system, equipment and storage medium based on supply chain system | |
Lawrencia et al. | Fraud detection decision support system for Indonesian financial institution | |
KR20140029994A (en) | Method for verifying sutability of loss given default/exposure at default |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |