Method and apparatus for multi-dimensional analysis of data for the insurance industry

Abstract

The application provides a data multidimensional analysis method and device for the insurance industry, which comprises the following steps: acquiring all dimensional information for risk analysis of the insurance industry; acquiring all index information for measuring risk of the insurance industry; associating the dimensional information, the index information and entity table relations in a database through a knowledge graph, and constructing a fact table based on the association result; establishing a metadata model based on the dimensional information, the index information and the fact table, wherein the metadata model is an entity relation model; inputting dimensional information and index information to be analyzed for risk into the metadata model, and obtaining a fact table with minimum query cost; and obtaining an optimal query path meeting a query condition and a query result corresponding to the optimal query path based on the fact table with minimum query cost. The method can realize fast response of big data multidimensional analysis and improve user experience.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method and apparatus for multidimensional data analysis in the insurance industry. Background Technology

[0002] Business quality monitoring and risk analysis are key focuses of statistical analysis in the insurance industry. Providing timely and flexible, fine-grained analytical perspectives, and calculating multi-dimensional risk indicators such as loss ratios, premium rates, and claim rates in real time for risk warning, has become a key area of ​​exploration for insurance companies' big data application scenarios. Current technologies for multidimensional analysis suffer from the following problems: First, datasets are independent of each other, with inconsistencies in dimensions and indicator definitions. Second, they cannot support flexible multidimensional analysis of large datasets. When conducting risk pricing analysis, the insurance industry requires fine-grained data, often down to the policy level, with a long time span of 5-10 years, resulting in detailed data volumes reaching billions of records. Traditional multidimensional analysis cannot meet the needs of flexible multidimensional analysis based on such large datasets.

[0003] Based on the above solution, it is necessary to establish a model with unified dimensional indicators and utilize distributed computing technology to meet the need for flexible and multidimensional analysis of risk-related indicators in the insurance industry. Summary of the Invention

[0004] This invention provides a method and apparatus for multidimensional data analysis in the insurance industry, which can be used to perform rapid statistical analysis and more comprehensive monitoring of insurance business risks.

[0005] This invention provides a method for multidimensional data analysis in the insurance industry, comprising:

[0006] Obtain all dimensional information for risk analysis in the insurance industry, including dimensional categories and the corresponding dimensional hierarchy information;

[0007] Obtain all risk measurement indicator information in the insurance industry, including indicator categories and corresponding indicator codes for those categories;

[0008] The dimensional information, indicator information and the relationship between them and the entity tables in the database are associated through a knowledge graph, and a fact table is constructed based on the association results;

[0009] Based on the dimensional information, indicator information, and fact table, a metadata model is established, which is an entity relationship model.

[0010] The dimension and indicator information to be analyzed are input into the metadata model to obtain the fact table with the lowest query cost; based on the fact table with the lowest query cost, the optimal query path that meets the query conditions and the query results corresponding to the optimal query path are obtained.

[0011] According to the multidimensional data analysis method for the insurance industry provided by this invention, before acquiring all dimensional information for risk analysis in the insurance industry, the dimensional information includes dimensional categories and corresponding dimensional hierarchy information, the method includes:

[0012] The dimension categories in the database are uniformly coded, with each dimension category corresponding to a unique code;

[0013] The dimension category and dimension category code are stored in a dimension list, which also includes the database table name corresponding to the dimension, the fact table name associated with the dimension, and the dimension description.

[0014] The dimensions in the database are uniformly encoded, with each dimension corresponding to a unique code;

[0015] The dimension level and its encoding are stored in a level list, which also includes the database table name corresponding to the dimension level and the ID of the level list.

[0016] According to the multidimensional data analysis method for the insurance industry provided by this invention, all indicator information for measuring risk in the insurance industry is obtained. Before the indicator information includes indicator categories and corresponding indicator code information, it includes:

[0017] The indicator categories in the database are uniformly coded, with each indicator category corresponding to a unique code;

[0018] The indicator category and indicator category code are stored in the indicator list. The indicator list also includes the commonly used database table name of the indicator, the indicator unit, whether the indicator is calculated, and whether it is a point-in-time indicator.

[0019] According to the multidimensional data analysis method for the insurance industry provided by this invention, the dimensional information, indicator information, and entity table relationships in the database are associated through a knowledge graph. Based on the association results, a fact table is constructed, including:

[0020] The dimension list and hierarchy list corresponding to the dimension information, the indicator list corresponding to the indicator information, and the entity table in the database are combined in different ways and stored in the corresponding fact table.

[0021] The fact table also includes the fact table query cost, the fact table access IP and port; the fact table query cost is calculated based on the number of fact table records, the fact table skewness, and the performance of the machine where the fact table is located.

[0022] According to the multidimensional data analysis method for the insurance industry provided by this invention, the dimensional information and indicator information to be analyzed for risk are input into the metadata model to obtain a fact table with the lowest query cost, including:

[0023] Find the minimum cost fact table in the metadata model that meets the aggregation conditions of the dimensional metrics.

[0024] According to the multidimensional data analysis method for the insurance industry provided by this invention, based on the fact table with the lowest query cost, the optimal query path that satisfies the query conditions and the query results corresponding to the optimal query path are obtained, including:

[0025] The fact table with the lowest query cost is used as the entity table for the query, data statistics are performed, and multidimensional query results are returned.

[0026] The present invention also provides a data multidimensional analysis device for the insurance industry, comprising:

[0027] The acquisition module is used to acquire dimensional information for risk analysis in the insurance industry. The dimensional information includes dimensional categories and dimensional hierarchy information corresponding to the dimensional categories.

[0028] The acquisition module is also used to acquire all indicator information for measuring risk in the insurance industry, including indicator categories and indicator code information corresponding to the indicator categories;

[0029] The construction module is used to associate the dimensional information, indicator information and entity table relationships in the database through a knowledge graph, and construct a fact table based on the association results;

[0030] The construction module is also used to establish a metadata model based on the dimension information, indicator information and the fact table, wherein the metadata model is an entity relationship model;

[0031] The determination module is used to input the dimension information and indicator information to be analyzed into the metadata model to obtain the fact table with the lowest query cost;

[0032] The determining module is further configured to obtain the optimal query path that satisfies the query conditions and the query result corresponding to the optimal query path based on the fact table with the lowest query cost.

[0033] According to the data multidimensional analysis device for the insurance industry provided by the present invention, the determining module further includes:

[0034] The multidimensional analysis engine module is used to input the dimensional and indicator information to be analyzed into the metadata model to obtain multidimensional analysis data query results.

[0035] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the multidimensional data analysis method for the insurance industry as described above.

[0036] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the data multidimensional analysis method for the insurance industry as described above.

[0037] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements a multidimensional data analysis method for the insurance industry as described above.

[0038] This invention provides a method and apparatus for multidimensional data analysis in the insurance industry. Employing distributed technology, it constructs a metadata model for dimensional indicator analysis within the insurance industry. This model uses entity-relationship modeling to establish multidimensional analysis metadata for all business analysis dimensions, risk-related indicators, dimension tables, and fact tables. Based on this metadata, the multidimensional analysis engine analyzes user queries, finds the optimal query path, dynamically generates query language, and returns query results. This enables rapid response in big data multidimensional analysis, providing underlying data support for business quality monitoring and risk management in the insurance industry, and offering technical assurance for rapid multidimensional queries. Furthermore, this invention employs a distributed microservice architecture, providing multidimensional analysis queries through API interfaces. This not only distributes query pressure but also provides a unified interface service standard for various client requests. By applying distributed columnar databases, distributed caching technology, and distributed microservices to multidimensional analysis queries, it improves data access efficiency, query efficiency, and user experience with minimal storage usage. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0040] Figure 1 A flowchart illustrating a multidimensional data analysis method for the insurance industry provided in an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram of the structure of a data multidimensional analysis device for the insurance industry provided in an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0044] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms, unless the context clearly indicates otherwise. “Multiple” generally includes at least two, but does not exclude the inclusion of at least one.

[0045] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0046] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a product or system comprising a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a product or system. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the product or system that includes said element.

[0047] This invention provides a multidimensional data analysis method for the insurance industry, which utilizes distributed technology to perform multidimensional data analysis for rapid statistical analysis and more comprehensive monitoring of insurance business risks.

[0048] The following is combined Figures 1-3 This invention describes a method and apparatus for multidimensional data analysis in the insurance industry.

[0049] Figure 1 This is a flowchart illustrating a multidimensional data analysis method for the insurance industry provided in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0050] Step 101: Obtain all dimensional information for risk analysis in the insurance industry, including dimensional categories and the corresponding dimensional hierarchy information;

[0051] Step 102: Obtain all risk measurement indicator information in the insurance industry, including indicator categories and indicator code information corresponding to the indicator categories;

[0052] Step 103: Associate the dimensional information, indicator information and entity table relationships in the database through a knowledge graph, and construct a fact table based on the association results;

[0053] Step 104: Based on the dimensional information, indicator information, and fact table, establish a metadata model, which is an entity-relationship model;

[0054] Step 105: Input the dimension information and indicator information to be analyzed into the metadata model to obtain the fact table with the lowest query cost; based on the fact table with the lowest query cost, obtain the optimal query path that meets the query conditions and the query results corresponding to the optimal query path.

[0055] The above steps will be explained in detail below with reference to specific embodiments.

[0056] Step 101: Obtain all dimensional information for risk analysis in the insurance industry, including dimensional categories and the corresponding dimensional hierarchy information;

[0057] In this step, information on multiple thematic dimensions of the insurance industry is obtained, including but not limited to: information on auto insurance (vehicle, person, product, insurance method, policy), information on agricultural insurance (classification of insured objects, type of breeding, product classification, sales channels), and information on property insurance (industry, inventory, institutions, products, channels).

[0058] Specifically, insurance industry risks are related to many dimensions of information. For example, auto insurance risks can be analyzed from multiple dimensions such as the issuing institution, product, sales channel, policyholder's age, vehicle age, number of claims, vehicle registration score, vehicle price, and vehicle model / series. Property insurance risks can be analyzed from multiple dimensions such as the institution, product, sales channel, inventory, industry, insured's address, building structure, and scenario. In practical applications, users will arbitrarily select several dimension combinations, add some constraints, and perform analysis. The traditional approach is to exhaustively list dimension combinations, pre-aggregate the data, and then statistically analyze the data if the user's query dimension information is among the pre-prepared dimension combinations. This invention stores all dimension information and dimension hierarchy information as dimension metadata in the database. The user's query dimension information can be any combination of dimensions. If the combined dimension data does not exist in the pre-processed dataset, the system automatically processes and aggregates the data from the list dataset to satisfy the user's data analysis of any combination of dimensions.

[0059] The dimensional information is encoded using both dimensional and hierarchical encoding methods. Dimensional information with different names but identical substance is identified. Unified dimensional information and hierarchical metadata are stored in the database using a two-dimensional table. Dimensions are encoded using a unified method, such as organization dimension code A001A, product dimension code A002A, and target dimension code A003A. Hierarchical dimensions are encoded using a unified method, such as provincial organization code L001L, municipal organization code L002L, county branch organization code L003L, product insurance type one L004L, product insurance type two L005L, etc. All dimensional information is stored in a single two-dimensional table, which includes the dimension name, dimension code, corresponding database table name, field names used in the associated fact table, and dimension description. All hierarchical information is also stored in a single two-dimensional table, which includes the hierarchical name, hierarchical code, corresponding database table field names, and the ID of the dimension table to which the hierarchical belongs.

[0060] Step 102: Obtain all risk measurement indicator information in the insurance industry, including indicator categories and indicator code information corresponding to the indicator categories;

[0061] In this step, after obtaining all dimensions and levels of information for risk analysis in the insurance industry, we will acquire all the indicators used by the insurance industry to measure risk. These indicators include, but are not limited to, premium income, earned premiums, claimed claims, unclaimed claims, loss ratio, incident rate, and number of valid claims reported. These indicators will be uniformly coded and identified, for example, premium income M001M, earned premiums M002M, etc. For example, all indicator information can be stored in a two-dimensional table. The table content includes the indicator code, indicator name, commonly used database table field names, indicator unit, whether the indicator is calculated, and whether it is a point-in-time indicator.

[0062] Specifically, the insurance industry uses numerous risk measurement indicators, such as loss ratio, claim rate, earned premiums, claimed claims, unclaimed claims, earned sum assured, and loss rate. Users conduct risk analysis by combining the dimensional information obtained in step 101 with the indicator information of this stage. Traditionally, this involves exhaustively listing dimensions and indicator combinations, pre-aggregating data, and then calculating the data based on the user's query for dimensional indicator information within the pre-prepared combinations. In this invention, all indicator information is stored as indicator metadata in the database. User queries can be any combination of dimensional indicators. If the combined data is not present in the pre-processed dataset, the system automatically processes and aggregates the data from the list set to satisfy the user's data analysis for any combination of dimensions. Indicator information is uniformly coded and identified, such as premium income M001M, earned premiums M002M, etc. All indicator information is stored in a two-dimensional table, the table content including indicator code, indicator name, commonly used database table field names, indicator unit, whether the indicator is calculated, and whether it is a point-in-time indicator, etc.

[0063] Step 103: Associate the dimensional information, indicator information and entity table relationships in the database through a knowledge graph, and construct a fact table based on the association results;

[0064] In this step, after obtaining all dimensions and levels of information for risk analysis in the insurance industry and all indicators for measuring risk in the insurance industry, a fact table data containing several dimensions and levels and several combinations of indicators is established.

[0065] Step 104: Based on the dimensional information, indicator information, and fact table, establish a metadata model, which is an entity-relationship model;

[0066] In this step, all fact table metadata is stored in a two-dimensional table. The table content includes the fact table name, fact table description, list of dimensions, list of levels, list of metrics, query cost, and access IP and port. Dimension, level, and metric lists are stored in the form of codes followed by commas. For example, if a fact table includes organizational, product, and channel dimensions, the dimension list would be stored as "A001A, A002A, A003A". All list codes are stored in coded order. The fact table query cost is calculated based on factors such as the number of fact table records, fact table skewness, and the performance of the machine hosting the fact table, thereby constructing the entity-relationship model, i.e., the metadata model.

[0067] Step 105: Input the dimension information and indicator information to be analyzed into the metadata model to obtain the fact table with the lowest query cost; based on the fact table with the lowest query cost, obtain the optimal query path that meets the query conditions and the query results corresponding to the optimal query path.

[0068] In this step, a unified dimensional metric metadata model is used to retrieve the fact table with the lowest cost that can provide the required set of dimensions and metrics for the query. The metadata information of the fact table is used to automatically generate a query language, which is then submitted to the server API where the fact table resides. The query language is executed, and the query results are returned. The combination of dimensions and metrics for each query request is recorded in the background. The model periodically generates a new fact table based on the number of requests, thus ensuring that frequently queried combinations of dimensions and metrics are supported by a fact table with lower cost, achieving adaptive functionality.

[0069] Specifically, the dimensional information and indicator information mentioned here can be used as input variables for multidimensional analysis. The dimensional combination information, hierarchical combination information, filtering conditions, and indicator combination information are input into the multidimensional analysis engine. The engine returns the optimal query statement, executes the statement, and returns the query results. The dimensional combination information can be any combination of dimensions related to insurance industry risk, the hierarchical combination information can be any combination of dimensions and hierarchical levels related to insurance industry risk, and the indicator combination information can be any combination of indicators related to insurance industry risk.

[0070] Furthermore, the aforementioned dimension combinations, dimension hierarchy combinations, and indicator combinations are converted into internal coded identifiers. The fact table information that satisfies the above combination conditions with the lowest query cost is found in the metadata model. By analyzing the fact table metadata, a query language is automatically generated, the query statement is executed on the corresponding server, and the query results are returned.

[0071] It is worth noting that if no matching fact table is found, a summary fact table that meets the query conditions is generated based on the list-level metadata model, and the newly generated fact table is updated in the fact table metadata to meet the query requirements.

[0072] Furthermore, the system periodically checks the combinations of dimensions and metrics used by users, creating new fact tables with lower query costs to meet the query needs of frequently used dimension and metric combinations. These new fact tables are then appended to the metadata model to adaptively improve the efficiency of multidimensional analysis.

[0073] Understandably, metadata can be stored in a distributed cache database, fact table and dimension table data can be stored in a columnar distributed database, and recently accessed data results can also be stored in a distributed cache database to improve the efficiency of multidimensional analysis and retrieval.

[0074] Overall, the process begins by converting dimension combination information into dimension code combinations, sorting them, and separating them with commas and matching operators. Then, dimension hierarchy combination information is converted into dimension hierarchy code combinations, sorted, and separated with commas and matching operators. Next, indicator combination information is converted into indicator code combinations, sorted, and separated with commas and matching operators. The sorted dimension code combinations, dimension hierarchy code combinations, and indicator code combinations are then matched against the fact table metadata to find all fact tables that meet the criteria. The fact table with the lowest query cost is selected as the fact table used in the query language. Finally, by automatically parsing the relationship between the fact table metadata and the dimension table metadata, dimension hierarchy table metadata, and indicator table metadata, a query statement is automatically generated according to standard query language specifications. The query statement is then submitted to the server where the fact tables reside for execution, returning query results in JSON format for front-end display.

[0075] If no matching fact table is found in the above steps, the fact table is automatically created from the list layer, and the fact table metadata is maintained in the metadata to participate in subsequent analysis engine calculations.

[0076] In other words, when conducting multidimensional risk analysis in the insurance industry, not only is all dimensional information and dimensional hierarchy combination information obtained, but also indicator combination information. Specifically, the dimensional combination information, dimensional hierarchy combination information, and indicator combination information are input into the established insurance industry dimensional indicator metadata model to obtain fact table metadata information that supports the dimensional indicators. Based on the fact table metadata information, a query language is automatically generated to complete the query operation and return the query results. Based on the above solution, by establishing an adaptive insurance industry dimensional indicator metadata model, various analysis systems and data marts existing within the company are unified into a single dimensional indicator metadata model. Utilizing the computing power of the hardware resources distributed across these analysis systems and data marts, a unified standard for dimensional indicators in the insurance industry is formed, significantly improving the efficiency of multidimensional analysis.

[0077] The following describes the data multidimensional analysis device for the insurance industry provided by the present invention. The data multidimensional analysis device for the insurance industry described below and the data multidimensional analysis method for the insurance industry described above can be referred to in correspondence.

[0078] Figure 2 This is a schematic diagram of the structure of a data multidimensional analysis device for the insurance industry provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the device includes:

[0079] The acquisition module 21 is used to acquire dimensional information for risk analysis in the insurance industry. The dimensional information includes dimensional categories and dimensional hierarchy information corresponding to the dimensional categories.

[0080] The acquisition module 21 is also used to acquire all indicator information for measuring risk in the insurance industry, including indicator categories and indicator code information corresponding to the indicator categories;

[0081] Module 22 is used to associate the dimension information, indicator information and entity table relationships in the database through a knowledge graph, and construct a fact table based on the association results;

[0082] The construction module 22 is further configured to establish a metadata model based on the dimension information, indicator information and the fact table, wherein the metadata model is an entity relationship model;

[0083] The determination module 23 is used to input the dimension information and indicator information to be analyzed into the metadata model to obtain the fact table with the lowest query cost;

[0084] The determining module 23 is further configured to obtain the optimal query path that satisfies the query conditions and the query result corresponding to the optimal query path based on the fact table with the lowest query cost.

[0085] Optionally, module 21 is used specifically for:

[0086] The dimension categories in the database are uniformly coded, with each dimension category corresponding to a unique code;

[0087] The dimension category and dimension category code are stored in a dimension list, which also includes the database table name corresponding to the dimension, the fact table name associated with the dimension, and the dimension description.

[0088] The dimensions in the database are uniformly encoded, with each dimension corresponding to a unique code;

[0089] The dimension level and its encoding are stored in a level list, which also includes the database table name corresponding to the dimension level and the ID of the level list.

[0090] Optionally, module 21 is further used for:

[0091] The indicator categories in the database are uniformly coded, with each indicator category corresponding to a unique code;

[0092] The indicator category and indicator category code are stored in the indicator list. The indicator list also includes the commonly used database table name of the indicator, the indicator unit, whether the indicator is calculated, and whether it is a point-in-time indicator.

[0093] Optionally, module 22 is constructed specifically for:

[0094] The dimension list and hierarchy list corresponding to the dimension information, the indicator list corresponding to the indicator information, and the entity table in the database are combined in different ways and stored in the corresponding fact table.

[0095] The fact table also includes the fact table query cost, the fact table access IP and port; the fact table query cost is calculated based on the number of fact table records, the fact table skewness, and the performance of the machine where the fact table is located.

[0096] Optionally, module 23 is defined, specifically for:

[0097] Find the minimum cost fact table in the metadata model that meets the aggregation conditions of the dimensional metrics.

[0098] Optionally, module 23 is defined, specifically for:

[0099] The fact table will serve as the entity table for the query, performing data statistics and returning multidimensional analysis results.

[0100] The apparatus of this embodiment can be used to execute the method of any of the aforementioned embodiments of the electronic device side method. Its specific implementation process and technical effects are similar to those of the electronic device side method embodiments. For details, please refer to the detailed description in the electronic device side method embodiments, which will not be repeated here.

[0101] The multidimensional data analysis method and apparatus for the insurance industry provided in this invention adopts distributed technology, and for the first time applies distributed columnar databases, distributed caching technology, and distributed microservices to multidimensional analysis queries, improving query efficiency and enhancing user experience. It establishes multidimensional analysis metadata using entity-relationship modeling for all business analysis dimensions and risk-related indicators, dimension tables, and fact tables. For dimension information, it focuses on business significance, thus setting query-related system attributes such as sharding, access control, and data segmentation. For indicator information, it focuses on business meaning, thus configuring query-related system attributes such as unit, update frequency, whether to count, whether to use point-in-time indicators, and physical storage location, covering the entire indicator set for risk analysis.

[0102] This invention also constructs a metadata model for analyzing dimensional indicators in the insurance industry, enabling rapid response to multidimensional big data analysis and providing underlying data support and multidimensional rapid query technology for business quality monitoring and risk management in the insurance industry.

[0103] This invention also utilizes a multidimensional analysis engine, employing an adaptive optimization model strategy. Based on multidimensional analysis metadata, it analyzes user queries, finds the optimal query path, dynamically generates query language, and returns query results. It adopts a distributed microservice architecture, providing multidimensional analysis queries through API interfaces, which not only distributes query pressure but also provides a unified interface service standard for various client requests. The multidimensional analysis engine autonomously optimizes the model based on user query habits, pre-aggregates data, and improves data access efficiency and enhances user experience with minimal storage usage.

[0104] Figure 3 This is a schematic diagram of the physical structure of an electronic device provided in an embodiment of the present invention, such as... Figure 3As shown, the electronic device may include: a processor 310, a communication interface 320, a memory 330, and a communication bus 340, wherein the processor 310, the communication interface 320, and the memory 330 communicate with each other through the communication bus 340. The processor 310 can call logical instructions in the memory 330 to execute a multidimensional data analysis method for the insurance industry. This method includes: acquiring all dimensional information for risk analysis in the insurance industry, the dimensional information including dimensional categories and corresponding dimensional hierarchy information; acquiring all indicator information for measuring risk in the insurance industry, the indicator information including indicator categories and corresponding indicator codes; associating the dimensional information, indicator information, and entity tables in the database using a knowledge graph, and constructing a fact table based on the association results; establishing a metadata model based on the dimensional information, indicator information, and the fact table, the metadata model being an entity-relationship model; inputting the dimensional information and indicator information to be analyzed into the metadata model to obtain a fact table with the lowest query cost; and obtaining the optimal query path that satisfies the query conditions and the query results corresponding to the optimal query path based on the fact table with the lowest query cost.

[0105] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0106] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the multi-dimensional data analysis method for the insurance industry provided by the above methods. The method includes: acquiring all dimensional information for risk analysis in the insurance industry, the dimensional information including dimensional categories and corresponding dimensional hierarchy information; acquiring all indicator information for measuring risk in the insurance industry, the indicator information including indicator categories and corresponding indicator codes; associating the dimensional information, indicator information, and entity table relationships in the database through a knowledge graph, and constructing a fact table based on the association results; establishing a metadata model based on the dimensional information, indicator information, and fact table, the metadata model being an entity relationship model; inputting the dimensional information and indicator information to be analyzed into the metadata model to obtain a fact table with the lowest query cost; and obtaining the optimal query path that satisfies the query conditions and the query results corresponding to the optimal query path based on the fact table with the lowest query cost.

[0107] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the multi-dimensional data analysis method for the insurance industry provided by the above methods. This method includes: acquiring all dimensional information for risk analysis in the insurance industry, the dimensional information including dimensional categories and corresponding dimensional hierarchy information; acquiring all indicator information for measuring risk in the insurance industry, the indicator information including indicator categories and corresponding indicator codes; associating the dimensional information, indicator information, and entity tables in a database through a knowledge graph, and constructing a fact table based on the association results; establishing a metadata model based on the dimensional information, indicator information, and the fact table, the metadata model being an entity-relationship model; inputting the dimensional information and indicator information to be analyzed into the metadata model to obtain a fact table with the lowest query cost; and obtaining the optimal query path that satisfies the query conditions and the query results corresponding to the optimal query path based on the fact table with the lowest query cost.

[0108] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0109] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0110] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for multi-dimensional analysis of data for the insurance industry, characterized in that, include: Obtain all dimensional information for risk analysis in the insurance industry, including dimensional categories and the corresponding dimensional hierarchy information; Obtain all risk measurement indicator information in the insurance industry, including indicator categories and corresponding indicator codes for those categories; The dimensional information, indicator information and the relationship between them and the entity tables in the database are associated through a knowledge graph, and a fact table is constructed based on the association results; Based on the aforementioned dimension information, indicator information, and fact tables, a metadata model is established that includes uniformly managed dimension categories and their hierarchical levels, indicator categories and their codes, and the corresponding relationships of the fact tables. This metadata model is an entity-relationship model. Each fact table has a corresponding fact table description. When establishing the metadata model, the fact table also includes a fact table query cost, which is calculated based on the number of fact table records, the fact table skewness, and the performance of the machine where the fact table is located. Input the dimension and indicator information to be analyzed into the metadata model to obtain a fact table that covers the query content and has the lowest query cost; Based on the fact table with the lowest query cost, the optimal query path that satisfies the query conditions and the query results corresponding to the optimal query path are obtained. The method also includes: recording the combination of dimensional metrics and the number of requests for each query request in the background; Based on the number of requests, a new fact table with lower query cost is automatically generated for frequently used dimension metrics combinations, and the new fact table is appended to the metadata model.

2. The method for multi-dimensional analysis of data for the insurance industry as claimed in claim 1, wherein, Before acquiring all dimensional information for risk analysis in the insurance industry, including dimensional categories and corresponding dimensional hierarchy information, the following information is included: The dimension categories in the database are uniformly coded, with each dimension category corresponding to a unique code; The dimension category and dimension category code are stored in a dimension list, which also includes the database table name corresponding to the dimension, the fact table name associated with the dimension, and the dimension description. The dimensions in the database are uniformly encoded, with each dimension corresponding to a unique code; The dimension level and its encoding are stored in a level list, which also includes the database table name corresponding to the dimension level and the ID of the level list.

3. The method for multi-dimensional analysis of data for the insurance industry as claimed in claim 1 wherein, Before acquiring all risk measurement indicator information in the insurance industry, including indicator categories and corresponding indicator codes for those categories, the following information is required: The indicator categories in the database are uniformly coded, with each indicator category corresponding to a unique code; The indicator category and indicator category code are stored in the indicator list. The indicator list also includes the commonly used database table name of the indicator, the indicator unit, whether the indicator is calculated, and whether it is a point-in-time indicator.

4. The method for multi-dimensional analysis of data for the insurance industry as claimed in claim 1, wherein, The dimensional information, indicator information, and relationships between them and entity tables in the database are linked using a knowledge graph. Based on the association results, a fact table is constructed, including: The dimension list and hierarchy list corresponding to the dimension information, the indicator list corresponding to the indicator information, and the entity table in the database are combined in different ways and stored in the corresponding fact table. The fact table also includes the fact table access IP and port.

5. The multidimensional data analysis method for the insurance industry according to claim 1, characterized in that, By inputting the dimensional and indicator information to be analyzed into the metadata model, a fact table with the lowest query cost is obtained, including: Find the minimum cost fact table in the metadata model that meets the aggregation conditions of the dimensional metrics.

6. The multidimensional data analysis method for the insurance industry according to claim 1, characterized in that, Based on the fact table with the lowest query cost, the optimal query path that satisfies the query conditions and the query results corresponding to the optimal query path are obtained, including: The fact table serves as the entity table for the query, performs data statistics, and returns multidimensional analysis results.

7. A data multidimensional analysis device for the insurance industry, characterized in that, include: The acquisition module is used to acquire all dimensional information for risk analysis in the insurance industry. The dimensional information includes dimensional categories and the dimensional hierarchy information corresponding to the dimensional categories. The acquisition module is also used to acquire all indicator information for measuring risk in the insurance industry, including indicator categories and indicator code information corresponding to the indicator categories; The construction module is used to associate the dimensional information, indicator information and entity table relationships in the database through a knowledge graph, and construct a fact table based on the association results; The construction module is further configured to establish a metadata model based on the dimension information, indicator information, and fact table. This metadata model includes uniformly managed dimension categories and their hierarchical levels, indicator categories and their codes, and the corresponding relationships of the fact table. The metadata model is an entity-relationship model. Each fact table has a corresponding fact table description. When establishing the metadata model, the fact table also includes a fact table query cost, which is calculated based on the number of fact table records, the fact table skewness, and the performance of the machine where the fact table is located. The building module is also used to: record the combination of dimension metrics and the number of requests for each query request in the background; periodically generate new fact tables with lower query costs for frequently used dimension metrics combinations based on the number of requests, and append the new fact tables to the metadata model; The determination module is used to input the dimension information and indicator information to be analyzed into the metadata model to obtain a fact table that covers the query content and has the lowest query cost; The determining module is further configured to obtain the optimal query path that satisfies the query conditions and the query result corresponding to the optimal query path based on the fact table with the lowest query cost.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the data multidimensional analysis method for the insurance industry as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the data multidimensional analysis method for the insurance industry as described in any one of claims 1 to 6.

Images