Insurance policy data compression method and device, electronic equipment and storage medium

Abstract

The invention provides an insurance policy data compression method and device, electronic equipment and a storage medium, and is applied to the fields of finance and medical treatment. The insurance policy data compression method provided by the invention comprises the steps of obtaining to-be-compressed insurance policy data, and extracting feature data of the to-be-compressed insurance policy data, the feature data comprising data distribution feature data, correlation feature data and volatility feature data; determining a service demand according to an accounting rule and a service target of IFRS17, and determining a compression strategy according to the feature data and the service demand; and generating an SQL statement according to the compression strategy, and performing compression operation on the to-be-compressed insurance policy data according to the SQL statement to obtain compressed data. According to the insurance policy data compression method, the development cost of the compression scheme can be reduced, and the compression efficiency and the compression accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular relates to a policy data compression method and device, electronic equipment and storage medium. BACKGROUND

[0002] In the medical field and the financial field, there are data cleaning and compression scenarios. In the implementation process of IFRS17 international accounting standards, data cleaning and compression are also one of the key links. However, the existing compression scheme is usually embedded in the code in a fixed way, and cannot be flexibly adjusted according to data characteristics and business needs. When the accounting rules of IFRS17 change or the data scale expands, the code needs to be manually modified and redeployed, which has a high development cost. In addition, the existing compression scheme lacks intelligent analysis and optimization capabilities, and cannot dynamically adjust the compression strategy according to the data characteristics, resulting in low compression efficiency and compression accuracy. SUMMARY

[0003] The present application aims to provide a policy data compression method, device, electronic equipment and storage medium to solve the technical problems of high development cost, low compression efficiency and low compression accuracy of the existing compression scheme in the prior art.

[0004] The technical scheme of the present application is as follows, a policy data compression method is provided, comprising:

[0005] Obtaining the compressed policy data, extracting the feature data of the compressed policy data, the feature data including data distribution feature data, correlation feature data and volatility feature data;

[0006] According to the accounting rules of IFRS17 and the business target, the business demand is determined, and the compression strategy is determined according to the feature data and the business demand;

[0007] According to the compression strategy, a SQL statement is generated, and a compression operation is performed on the compressed policy data according to the SQL statement, to obtain compressed data.

[0008] Further, according to the SQL statement, the compressed policy data is executed to obtain the compressed data, comprising:

[0009] The SQL statement is submitted to the database, and the compressed policy data is executed to obtain the compressed data, the compression operation including grouping compression, interval compression or statistical compression.

[0010] Further, according to the SQL statement, the compressed policy data is executed to obtain the compressed data, comprising:

[0011] The task of the SQL statement is decomposed into multiple subtasks, the multiple subtasks are respectively allocated to different CPU cores and executed in parallel, after all the subtasks are completed, the results of each subtask are merged to complete the compression operation on the to-be-compressed insurance data, and compressed data is obtained.

[0012] Further, the insurance data compression method further comprises;

[0013] If the compressed data does not meet the business requirement, at least one of the compression algorithm type, the compression algorithm parameter and the compression rule in the compression strategy is adjusted, the compression operation is performed again on the to-be-compressed insurance data according to the adjusted compression strategy, and the business requirement includes classification and / or aggregation of insurance data in a time period and / or a region;

[0014] If the compression ratio of the to-be-compressed insurance data and the compressed data is less than a preset compression threshold, or the time of performing the compression operation is greater than a preset time threshold, the compression algorithm type in the compression strategy is adjusted, and the compression operation is performed again on the to-be-compressed insurance data according to the adjusted compression strategy.

[0015] Further, determining the compression strategy according to the feature data and the business requirement comprises: selecting the compression strategy from a preset configuration table according to the feature data and the business requirement, and the compression strategy comprises a compression algorithm type, a compression algorithm parameter and a compression rule.

[0016] Further, generating the SQL statement according to the compression strategy comprises: generating the SQL statement according to the compression algorithm type, the compression algorithm parameter and the compression rule.

[0017] Further, after determining the compression strategy according to the feature data and the business requirement, before performing the compression operation on the to-be-compressed insurance data according to the SQL statement, the method further comprises performing a data cleaning operation on the to-be-compressed insurance data, and the data cleaning operation comprises at least one of removing duplicate data, filling missing values, correcting error data and standardizing data format.

[0018] Another technical scheme of the application provides an insurance data compression device, comprising a feature extraction module, a compression strategy determination module and a compression operation module.

[0019] The feature extraction module is configured to obtain to-be-compressed insurance data, and extract feature data of the to-be-compressed insurance data, wherein the feature data comprises data distribution feature data, correlation feature data and volatility feature data.

[0020] The compression strategy determination module is configured to determine business requirements according to accounting rules and business objectives of IFRS17, and determine a compression strategy according to the feature data and the business requirements.

[0021] The compression operation module is configured to generate an SQL statement according to the compression strategy, and perform a compression operation on the to-be-compressed insurance policy data according to the SQL statement to obtain compressed data.

[0022] Another technical solution of the present application is as follows: an electronic device is provided, comprising a memory and a processor, the memory stores a computer program executable by the processor, and the processor implements the insurance policy data compression method according to any one of the above technical solutions when executing the computer program.

[0023] Another technical solution of the present application is as follows: a computer readable storage medium is provided, the computer readable storage medium stores a computer program, and the computer program implements the insurance policy data compression method according to any one of the above technical solutions when executed by a processor.

[0024] The present application has the beneficial effects that: to-be-compressed insurance policy data is acquired, feature data of the to-be-compressed insurance policy data is extracted, the feature data comprises data distribution feature data, correlation feature data and volatility feature data; business requirements are determined according to accounting rules and business objectives of IFRS17, and a compression strategy is determined according to the feature data and the business requirements; an SQL statement is generated according to the compression strategy, and a compression operation is performed on the to-be-compressed insurance policy data according to the SQL statement to obtain compressed data; through the above technical solution, compression of insurance policy data can be realized, development cost of a compression scheme can be reduced, and compression efficiency and compression accuracy can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 A flowchart of an insurance policy data compression method provided by an embodiment of the present application is shown in the figure.

[0026] Figure 2 A structural diagram of an insurance policy data compression device provided by an embodiment of the present application is shown in the figure.

[0027] Figure 3 A structural diagram of an electronic device provided by an embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0028] With reference to the accompanying drawings, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0029] In the description of the present application, the words "first", "second", etc. are used only for the purpose of distinguishing the described embodiments, and cannot be understood as indicating or implying relative importance, nor indicating or implying sequence. In the present specification, the terms "comprise", "include", "have" and their variants mean "including but not limited to", unless otherwise specifically emphasized.

[0030] In the present application, the term "embodiment" means that the specific features, structures or characteristics described in the embodiment can be included in at least one embodiment of the present application. The phrase appears at various places in the specification does not necessarily mean the same embodiment, nor is it an independent or alternative embodiment to other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0031] Figure 1 is a flowchart of the policy data compression method of the embodiments of the present application. It should be noted that the policy data compression method of the present application is not limited to the flow sequence shown in Figure 1 , as shown in Figure 1 , the policy data compression method mainly includes the following steps:

[0032] S101, obtaining the to-be-compressed policy data, extracting the feature data of the to-be-compressed policy data, the feature data including data distribution feature data, correlation feature data and volatility feature data;

[0033] In some embodiments, before extracting the feature data of the to-be-compressed insurance policy data, preprocessing of the data can be required, including cleaning, standardization, normalization, and other operations, to ensure the quality and consistency of the data. Machine learning algorithms such as principal component analysis (PCA), autoencoder, and the like can be used to extract feature data from the to-be-compressed insurance policy data, which can include data distribution feature data, correlation feature data, and volatility feature data, reflecting the distribution, correlation, volatility, and other characteristics of the data. The extracted feature data can be analyzed to identify patterns and structures in the to-be-compressed insurance policy data, such as analyzing the distribution of data in different intervals, the correlation between variables, and the stability of data over time. As an example, a set of insurance policy data, including policy ID, effective date, region, amount, and other fields, can be extracted by machine learning algorithms to extract the following features: data distribution data, i.e., the distribution of policy amounts in different intervals, such as the proportion of low, medium, and high value policies; correlation data, i.e., the correlation between effective date and region, such as higher number of policies in certain regions during certain time periods; and volatility data, i.e., the volatility of amounts over time, such as greater volatility of policy amounts in certain months. Through these features, the structure of the data can be better understood, providing a basis for subsequent compression strategies.

[0034] S102, determining business requirements according to the accounting rules and business objectives of IFRS17, and determining a compression strategy according to the feature data and the business requirements;

[0035] In some embodiments, according to the accounting rules of IFRS17 (International Financial Reporting Standards), the business requirements for data compression (including cleaning) are clearly defined, and the insurance policy data needs to be classified and aggregated according to specific time periods and regions. Specifically, the objectives of data cleaning and compression are determined, such as improving data processing efficiency, ensuring data accuracy and consistency, and meeting reporting requirements. The business requirements can be refined into specific data processing tasks, such as cleaning invalid data, classifying and aggregating data, and generating reports. As an example, if the insurance policy data needs to be classified and aggregated according to different time periods and regions to meet the reporting requirements of IFRS17, the business requirement analysis can include the following steps: determining the classification basis, classifying the insurance policy data according to the effective date and region; setting aggregation rules, summing or averaging the amounts according to the classification results; setting precision and efficiency requirements to ensure that the precision of the classification and aggregation results reaches the pre-set precision threshold and the data processing efficiency reaches the pre-set efficiency threshold. Through these steps, the business requirements for data compression can be clearly defined, providing direction for subsequent strategy optimization.

[0036] In an optional implementation, the compression strategy is determined according to the feature data and the business requirement, including: selecting the compression strategy from a preset configuration table according to the feature data and the business requirement, the compression strategy including a compression algorithm type, a compression algorithm parameter, and a compression rule.

[0037] In some embodiments, a configuration table (preset configuration table) can be designed in the database to store the related parameters and rules of the compression algorithm, the fields of the configuration table including but not limited to: algorithm name (Algorithm Name), algorithm type (Algorithm Type), compression rule (Compression Rules), parameter configuration (Parameters), and priority (Priority). Among them, the algorithm name can include group compression, interval compression, Huffman coding, etc.; the algorithm type can include group compression, clustering compression, statistical compression, etc.; the compression rule can include grouping according to time range, grouping according to region, grouping according to amount range, etc.; the parameter configuration can include grouping field, grouping condition, aggregation function, etc.; the priority can include high priority, medium priority, and low priority. The interval compression algorithm can group and compress according to the interval range of the data. Huffman coding is a frequency-based statistical compression algorithm, which is suitable for reducing the storage space of text or symbol data. When determining the compression strategy, the corresponding compression algorithm type can be determined according to the priority of the algorithm.

[0038] As an example, the algorithm name is group compression, the algorithm type is group compression, the compression rule is grouping according to time range and region, the grouping field in the parameter configuration includes effective_date and region, the grouping condition includes effective_date >= '2025-01-01' AND effective_date <= '2025-12-31', the parameter configuration includes the aggregation function, and the priority is high priority. Through the above configuration, the policies can be grouped according to the effective date and region, and the amount of each group can be summed to complete the data cleaning and compression. Through the configuration table, the user can flexibly adjust the parameters and rules of the compression algorithm; adjusting the parameters and rules of the compression algorithm can include increasing the number of grouping fields, modifying the grouping condition, adjusting the aggregation function, changing the basis of grouping, adjusting the aggregation function, and modifying the data cleaning rule.

[0039] In some embodiments, the parameters of the compression algorithm can be determined or dynamically adjusted according to the extracted data features and the analyzed business requirements, for example, adjusting the grouping fields, grouping conditions, aggregation functions, etc. According to the changes in data features and business requirements, the rules in the compression strategy are determined, for example, changing the basis for grouping, adjusting the target and rules of compression, and then generating the optimal compression strategy to ensure that the effect of data compression reaches the best. As an example, assuming that the business requirements include determining the need to classify and aggregate the policy data according to the effective date and region, in the data feature extraction, it is found that the policy amount in some regions fluctuates greatly in a certain time period, based on this information, the following optimization can be performed: adjusting the grouping fields, i.e., adjusting the grouping fields from "effective date" and "region" to "effective date", "region", and "amount range"; optimizing the grouping conditions, i.e., adjusting the grouping conditions according to the fluctuation of the amount range, for example, dividing the amount range into three intervals of low, medium, and high; adjusting the aggregation function, i.e., adjusting the aggregation function from "sum" to "average" according to the business requirements, to better reflect the policy amount situation in different regions. Through these adjustments and optimizations, a compression strategy that better meets the business requirements and data features can be generated, thereby improving the efficiency and accuracy of data processing.

[0040] S103, generating an SQL statement according to the compression strategy, and performing a compression operation on the to-be-compressed policy data according to the SQL statement to obtain compressed data.

[0041] In some embodiments, the data in the configuration table is read from the preset database to obtain the currently effective compression algorithm parameters and rules, and then the logic of the compression algorithm is parsed. If the compression algorithm is a grouping compression algorithm, the grouping conditions and aggregation functions are generated according to the parameters; specifically, the parameters related to grouping compression can be read from the configuration table, including grouping fields, grouping conditions, and aggregation functions. According to the read parameters, the specific logic of grouping compression is determined, for example, the grouping fields and aggregation functions are determined; based on the parsed logic, the specific SQL statement part is generated. As an example, assuming that the configuration table contains the following parameters, grouping fields: policy_id, grouping conditions: effective_date>='2025-01-01'ANDeffective_date<='2025-12-31', aggregation functions: SUM(amount), the parsed logic is: group by policy_id. After grouping, the amount field is summed. The generated SQL statement part is: WHERE effective_date>='2025-01-01'AND effective_date<='2025-12-31'GROUP BY policy_idHAVINGSUM(amount)AS total_amount.

[0042] In some embodiments, if the compression algorithm is interval compression, interval partitioning logic is generated according to the parameters. Specifically, parameters related to interval compression are read from the preset configuration table, including interval field, interval partitioning rule and processing logic; according to the read parameters, the specific logic of interval compression is determined, for example, the interval field and the partition range are determined; based on the parsed logic, the specific SQL statement part is generated. As an example, assume that the preset configuration table contains the following parameters, interval field: amount, low interval: amount<1000, medium interval: 1000<=amount<5000, high interval: amount>=5000; processing logic: count the data in each interval. The parsed logic includes: interval partitioning according to the amount field, and counting the data in each interval. The generated SQL statement part is: CASE WHEN amount<1000 THEN 'Low' WHEN amount>=1000 AND amount<5000 THEN 'Medium' ELSE 'High' END AS amount_range, COUNT(*) AS count_per_range.

[0043] In an optional implementation, the SQL statement is generated according to the compression strategy, including: generating the SQL statement according to the compression algorithm type, the compression algorithm parameters and the compression rule.

[0044] In some embodiments, the SQL statement is generated according to the compression algorithm type, the compression algorithm parameters, and the compression rules. Specifically, the currently effective compression algorithm type, parameters, and rules are obtained from a configuration table. According to the read parameters, the specific implementation logic of the compression algorithm is parsed, such as grouping fields, grouping conditions, aggregation functions, etc. The SQL statement is dynamically generated, and the corresponding compression SQL statement is dynamically generated according to the parsing result. The SQL statement can include selecting the required fields, specifying the data source table, adding filtering conditions, specifying the grouping field, applying the aggregation function, and other necessary SQL clauses. As an example, suppose the parsing result includes the algorithm name: Group Compression, the grouping field: policy_id, the grouping condition: effective_date>='2025-01-01' AND effective_date<='2025-12-31', and the aggregation function: SUM(amount). According to these parsing results, the dynamically generated compression SQL statement is: SELECT policy_id, SUM(amount) AS total_amount FROM policies WHERE effective_date>='2025-01-01' AND effective_date<='2025-12-31' GROUP BY policy_id.

[0045] According to the compression algorithm type, the compression algorithm parameters, and the compression rules, the SQL statement can be dynamically generated, which can flexibly adapt to different compression algorithms and business requirements without manual code modification, thereby improving the flexibility and maintainability of the system.

[0046] In an optional implementation, after determining the compression strategy according to the feature data and the business requirements, before performing the compression operation on the to-be-compressed policy data according to the SQL statement, the method further includes performing a data cleaning operation on the to-be-compressed policy data, and the data cleaning operation includes at least one of removing duplicate data, filling missing values, correcting erroneous data, and standardizing data formats.

[0047] In some embodiments, before performing the compression operation on the to-be-compressed insurance data according to the SQL statement, a data cleaning operation is further performed on the to-be-compressed insurance data, the data cleaning operation including at least one of removing duplicate data, filling in missing values, correcting erroneous data, and standardizing data formats; and then performing the compression operation on the cleaned to-be-compressed insurance data according to the SQL statement. Removing duplicate data can include identifying and deleting duplicate records through unique keys or combined keys; filling in missing values can include filling in missing data, common methods including mean filling, median filling, or using a machine learning model to predict missing values; correcting erroneous data can include identifying and correcting data that does not conform to business rules, such as invalid date formats or out-of-range numerical values; and standardizing data formats can include unifying different formats of data into a standard format, such as unifying dates into the YYYY-MM-DD format.

[0048] In an optional implementation, performing the compression operation on the to-be-compressed insurance data according to the SQL statement to obtain compressed data includes:

[0049] submitting the SQL statement to a database to perform the compression operation on the to-be-compressed insurance data to obtain compressed data, the compression operation including grouped compression, interval compression, or statistical compression.

[0050] In some embodiments, the compression operation aims to reduce data storage space while preserving important information. Grouped compression can include grouping data according to specific fields (such as time, region), and aggregating data within each group, such as summing, averaging, etc. Interval compression can include dividing data into intervals and processing data within each interval, such as counting the number of data within the interval or calculating statistical values within the interval. Statistical compression can include compressing data through statistical methods (such as statistical distribution, frequency analysis) to preserve key statistical indicators. These compression operations are implemented through the generation and execution of SQL statements, for example, cleaning operations can involve SQL statements that delete duplicate records, and compression operations can involve SQL statements that group and aggregate.

[0051] In an optional implementation, performing the compression operation on the to-be-compressed insurance data according to the SQL statement to obtain compressed data includes:

[0052] decomposing the task of the SQL statement into multiple sub-tasks, assigning the multiple sub-tasks to different CPU cores for parallel execution, and after all the sub-tasks are completed, merging the results of each sub-task to complete the compression operation on the to-be-compressed insurance data to obtain compressed data.

[0053] In some embodiments, during the compression operation on the to-be-compressed policy data, the execution overhead can be reduced by pre-compiling SQL statements. Specifically, when a SQL statement is executed for the first time, a database management system (DBMS) parses it and generates an execution plan, which describes in detail how to execute the SQL statement, including the way of accessing tables, the used indexes, the connection order, etc. The parsed SQL statement is compiled into intermediate code or machine code for subsequent fast execution; when the same SQL statement is executed subsequently, the pre-compiled execution plan and intermediate code are directly used without re-parsing and compiling, thereby reducing the execution time. Pre-compiling SQL statements is particularly suitable for scenarios where the same or similar SQL statements need to be frequently executed, such as the grouping aggregation operation in the data cleaning and compression process. By pre-compiling, the execution efficiency can be significantly improved, and the system load can be reduced.

[0054] In some embodiments, for large-scale data, the parallel execution of SQL statements can be compressed to improve the efficiency of compression processing. Specifically, a complex SQL task can be decomposed into multiple sub-tasks, each of which is responsible for processing a part of the data set, for example, in a grouping aggregation operation, the data set can be divided into multiple subsets according to the grouping field, and the aggregation processing is performed respectively; the decomposed sub-tasks are allocated to different CPU cores or nodes for parallel execution, and the database management system usually has a parallel query optimizer that can automatically determine how to decompose the task and allocate resources; after all the sub-tasks are completed, the results of the sub-tasks are merged to generate the final compressed data set. Through parallel execution, the multi-core processor and distributed computing resources can be fully utilized, the processing time can be significantly shortened, and the overall compression efficiency can be improved.

[0055] In an optional implementation, the policy data compression method further includes:

[0056] If the compressed data does not meet the business requirements, at least one of the compression algorithm type, the compression algorithm parameter, and the compression rule in the compression strategy is adjusted, and the compression operation is performed again on the to-be-compressed policy data according to the adjusted compression strategy, the business requirements including the classification and / or aggregation of policy data in a time period and / or a region;

[0057] If the compression ratio of the to-be-compressed policy data and the compressed data is less than a preset compression threshold, or the time of performing the compression operation is greater than a preset time threshold, the compression algorithm type in the compression strategy is adjusted, and the compression operation is performed again on the to-be-compressed policy data according to the adjusted compression strategy.

[0058] In some embodiments, the compression result can be recorded after the compression is completed, and the algorithm configuration can be optimized according to the result feedback. When the compressed data does not meet the expected target (does not meet the business requirements), it can be necessary to adjust the compression rules and / or parameters to optimize the result, including adjusting the grouping fields, modifying the grouping conditions, optimizing the aggregation functions, etc. Adjusting the grouping fields can include increasing or decreasing the grouping fields according to the data characteristics and business requirements, for example, if the current grouping fields are "effective_date" and "region", but the compression result fails to meet the expectations, the "amount_range" field can be added to improve the data subdivision level. Modify the grouping conditions to more accurately divide the data, for example, adjust the time range from "2025-01-01" to "2025-12-31" to a more detailed time interval. Optimizing the aggregation function includes selecting a more appropriate aggregation function according to the business requirements, for example, adjusting "SUM(amount)" to "AVG(amount)" to reflect the average policy amount in different regions. As an example, assume that the current compression rule is to group by "effective_date" and "region" and sum "amount", the compression result fails to meet the expected accuracy, it can be because the policy amount in some regions fluctuates greatly, at this time, the compression rule can be adjusted, "amount_range" is added as a grouping field, and the grouping conditions and aggregation functions are adjusted accordingly.

[0059] In some embodiments, if a certain compression algorithm performs poorly in a specific scenario, it can be dynamically switched to another algorithm. The judgment of poor performance of the algorithm is usually based on the following indicators: compression ratio, the ratio of the original data volume to the compressed data volume, the higher the compression ratio, the better the compression effect; processing time, the execution time of the compression operation, the shorter the processing time, the higher the efficiency; accuracy, whether the compressed data meets the business requirements, the higher the accuracy, the more reliable the result. If the performance of an algorithm in the above indicators is lower than the corresponding preset threshold, it can be judged that its performance is poor. Dynamic switching of algorithms is usually achieved through the following steps: monitoring and evaluation, real-time monitoring of algorithm execution and results during compression, evaluating whether the performance meets expectations; triggering the switching condition, if the evaluation result shows that the algorithm performs poorly, triggering the dynamic switching mechanism; selecting a replacement algorithm, selecting a more suitable algorithm according to data characteristics and business requirements, for example, if the grouping compression algorithm performs poorly, it can be switched to interval compression algorithm or statistical compression algorithm; updating the configuration, dynamically updating the parameters and rules in the configuration table to adapt to the new algorithm. When adjusting the compression algorithm type, compression algorithm parameters and compression rules in the compression strategy, it can be directly adjusted for compression operation, or the parameters in the preset configuration table can be adjusted, and the compression algorithm type, compression algorithm parameters and compression rules are obtained in the preset configuration table, and then the compression operation is performed.

[0060] The policy data compression method provided by the embodiment of the application can realize compression of policy data, reduce development cost of compression scheme, and improve compression efficiency and compression accuracy.

[0061] The policy data compression method provided by the embodiment of the application can dynamically optimize the compression strategy according to data characteristics and business requirements, improve compression effect and resource utilization; the change of the compression algorithm does not need to modify the code, but only needs to adjust the configuration in the database, which greatly shortens the development and deployment period; through the dynamic configuration management and closed-loop optimization mechanism, the workload of code maintenance and system restart is reduced, and the operation and maintenance cost is reduced; combined with intelligent analysis and dynamic optimization, the requirements of IFRS17 for data accuracy and consistency can be better met; the dynamic combination and adjustment of multiple compression algorithms can quickly adapt to changes in business requirements and updates of accounting rules.

[0062] The policy data compression method provided by the embodiment of the present application can be constructed based on artificial intelligence, and the related data is acquired and processed based on artificial intelligence technology, so as to realize unattended artificial intelligence policy data compression. The artificial intelligence (AI) is to use a digital computer or a machine controlled by a digital computer to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use the knowledge to obtain the best results.

[0063] The artificial intelligence basic technology generally includes technologies such as sensors, special artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operation / interaction system, mechatronics, etc. The artificial intelligence software technology mainly includes computer vision technology, robot technology, biometric identification technology, speech processing technology, natural language processing technology, and machine learning / deep learning, etc.

[0064] Figure 2 The structure diagram of the policy data compression device of the embodiment of the present application is shown in FIG. 1. Figure 2 As shown in FIG. 1, the policy data compression device 20 includes a feature extraction module 21, a compression strategy determination module 22 and a compression operation module 23.

[0065] The feature extraction module 21 is configured to acquire the to-be-compressed policy data, and extract feature data of the to-be-compressed policy data, wherein the feature data includes data distribution feature data, correlation feature data and volatility feature data.

[0066] The compression strategy determination module 22 is configured to determine a business requirement according to an accounting rule of IFRS17 and a business target, and determine a compression strategy according to the feature data and the business requirement.

[0067] The compression operation module 23 is configured to generate a SQL statement according to the compression strategy, perform a compression operation on the to-be-compressed policy data according to the SQL statement, and obtain compressed data.

[0068] In an optional implementation, the compression operation module 23 performs the compression operation on the to-be-compressed policy data according to the SQL statement to obtain the compressed data, including:

[0069] submitting the SQL statement to a database to perform the compression operation on the to-be-compressed policy data to obtain the compressed data, wherein the compression operation includes grouping compression, interval compression or statistical compression.

[0070] In an optional implementation, the compression operation module 23 performs the compression operation on the to-be-compressed policy data according to the SQL statement to obtain the compressed data, including:

[0071] The task of the SQL statement is decomposed into a plurality of sub-tasks, the plurality of sub-tasks are respectively allocated to different CPU cores and executed in parallel, after all the sub-tasks are completed, the results of each sub-task are merged to complete the compression operation on the to-be-compressed insurance data, and compressed data is obtained.

[0072] In an optional embodiment, the insurance data compression device 20 further comprises a compression optimization module; the compression optimization module is configured to adjust at least one of a compression algorithm type, a compression algorithm parameter and a compression rule in the compression strategy when the compressed data does not meet the business requirement, and perform a compression operation on the to-be-compressed insurance data according to the adjusted compression strategy, the business requirement including classification and / or aggregation of insurance data in a time period and / or a region.

[0073] The compression optimization module is further configured to adjust the compression algorithm type in the compression strategy when a compression ratio of the to-be-compressed insurance data and the compressed data is less than a preset compression threshold, or a time of performing the compression operation is greater than a preset time threshold, and perform the compression operation on the to-be-compressed insurance data according to the adjusted compression strategy.

[0074] In an optional embodiment, the compression strategy determination module 22 determines the compression strategy according to the feature data and the business requirement, including: selecting the compression strategy from a preset configuration table according to the feature data and the business requirement, the compression strategy including a compression algorithm type, a compression algorithm parameter and a compression rule.

[0075] In an optional embodiment, the compression operation module 23 generates a SQL statement according to the compression strategy, including: generating a SQL statement according to the compression algorithm type, the compression algorithm parameter and the compression rule.

[0076] In an optional embodiment, the insurance data compression device 20 further comprises a cleaning module, the cleaning module is configured to perform a data cleaning operation on the to-be-compressed insurance data before performing the compression operation on the to-be-compressed insurance data according to the SQL statement after determining the compression strategy according to the feature data and the business requirement, the data cleaning operation including at least one of removing duplicate data, filling missing values, correcting erroneous data and standardizing data formats.

[0077] Figure 3 is a structural schematic diagram of an electronic device of an embodiment of the present application. As shown in Figure 3 The electronic device 30 comprises a processor 31 and a memory 32 in communication connection with the processor 31.

[0078] The memory 32 stores program instructions for implementing the policy data compression method of any of the above embodiments.

[0079] The processor 31 is configured to execute the program instructions stored in the memory 32 to perform the policy data compression.

[0080] The processor 31 can also be referred to as a CPU (Central Processing Unit). The processor 31 can be an integrated circuit chip having a processing capability of signals. The processor 31 can also be a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application-Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0081] The storage medium of the embodiment of the present application stores program instructions capable of implementing all the above methods. The storage medium can be non-volatile or volatile. The program instructions can be stored in the above storage medium in the form of a software product, including a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in the various embodiments of the present application. The above storage medium includes a U disk, a mobile hard disk, a ROM (Read-Only Memory), a RAM (Random Access Memory), a magnetic disk or an optical disk, and various media capable of storing program codes, or a terminal device such as a computer, a server, a mobile phone, a tablet, etc.

[0082] In the several embodiments of the present application, it should be understood that the disclosed device and method can be implemented in other ways. For example, the device embodiments described above are only schematic. The division of the modules is only a logical function division. In actual implementation, another division mode can be adopted, for example, a plurality of modules or components can be combined or integrated into another system, or some features can be omitted or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed elements can be indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0083] In addition, each functional module in each embodiment of the present application can be integrated in one processing unit, or each module can exist physically separately, or two or more modules can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit. The above is only an embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent flow transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

[0084] The above is only an embodiment of the present application, and it should be pointed out that, for those skilled in the art, improvements can be made without departing from the inventive concept, but these are within the protection scope of the present application.

Claims

1. A method for compressing insurance policy data, characterized in that, include: Obtain the policy data to be compressed, and extract the feature data of the policy data to be compressed, including data distribution feature data, correlation feature data, and volatility feature data; Business needs are determined based on IFRS 17 accounting rules and business objectives, and a compression strategy is determined based on the characteristic data and the business needs. SQL statements are generated according to the compression strategy, and compression operations are performed on the policy data to be compressed according to the SQL statements to obtain compressed data.

2. The policy data compression method according to claim 1, characterized in that, The SQL statement is used to perform a compression operation on the policy data to be compressed, resulting in compressed data, including: The SQL statement is submitted to the database to perform a compression operation on the policy data to be compressed, resulting in compressed data. The compression operation includes group compression, interval compression, or statistical compression.

3. The policy data compression method according to claim 1, characterized in that, The SQL statement is used to perform a compression operation on the policy data to be compressed, resulting in compressed data, including: The SQL statement task is decomposed into multiple subtasks, which are then assigned to different CPU cores for parallel execution. After all subtasks are completed, the results of each subtask are merged to complete the compression operation of the policy data to be compressed, resulting in compressed data.

4. The policy data compression method according to claim 1, characterized in that, The policy data compression method further includes; If the compressed data does not meet the business requirements, at least one of the compression algorithm type, compression algorithm parameters and compression rules in the compression strategy is adjusted, and the compression operation is re-executed on the policy data to be compressed according to the adjusted compression strategy. The business requirements include the classification and / or summarization of policy data for time periods and / or regions. If the compression ratio between the policy data to be compressed and the compressed data is less than a preset compression threshold, or the time for performing the compression operation is greater than a preset time threshold, then the compression algorithm type in the compression strategy is adjusted, and the compression operation on the policy data to be compressed is performed again according to the adjusted compression strategy.

5. The policy data compression method according to claim 1, characterized in that, Determining a compression strategy based on the feature data and the business requirements includes: selecting the compression strategy from a preset configuration table based on the feature data and the business requirements, wherein the compression strategy includes compression algorithm type, compression algorithm parameters, and compression rules.

6. The policy data compression method according to claim 5, characterized in that, Generating SQL statements according to the compression strategy includes: generating SQL statements according to the compression algorithm type, the compression algorithm parameters, and the compression rules.

7. The policy data compression method according to claim 1, characterized in that, After determining the compression strategy based on the feature data and the business requirements, before performing the compression operation on the policy data to be compressed according to the SQL statement, the method further includes performing a data cleaning operation on the policy data to be compressed. The data cleaning operation includes at least one of removing duplicate data, filling in missing values, correcting erroneous data, and standardizing data format.

8. A policy data compression device, characterized in that, It includes a feature extraction module, a compression strategy determination module, and a compression operation module; The feature extraction module is used to acquire the policy data to be compressed and extract the feature data of the policy data to be compressed. The feature data includes data distribution feature data, correlation feature data and volatility feature data. The compression strategy determination module is used to determine business needs based on IFRS 17 accounting rules and business objectives, and to determine a compression strategy based on the feature data and the business needs. The compression operation module is used to generate an SQL statement according to the compression strategy, and perform a compression operation on the policy data to be compressed according to the SQL statement to obtain compressed data.

9. An electronic device, comprising a memory and a processor, wherein the memory stores a computer program executable by the processor, characterized in that, When the processor executes the computer program, it implements the policy data compression method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the policy data compression method as described in any one of claims 1 to 7.

Images