A financial data analysis method and system based on data analysis

Abstract

The application discloses a kind of financial data analysis method and system based on data analysis, it is related to data analysis technical field, the method includes: extracting the heterogeneous financial original data distributed in different business systems, according to the standardization alignment of preset attribute fingerprint characteristics to it wash and execute computing caliber, obtain standard financial data, according to financial analysis topological model, standard financial data are associated and mapped to multidimensional data theme and obtain multidimensional financial data, according to the logic operation of preset analysis index logic to multidimensional financial data, obtain key index data, establish the mapping association matrix between key index data and preset report template and fill in generation analysis chart, finally output financial analysis report.The application realizes the full-process automatic processing from heterogeneous data access to analysis report generation by the self-alignment of attribute fingerprint characteristics, the conflict verification of dimension topology and the dynamic mapping of index template, solves the problem of current automatic analysis interruption.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a financial data analysis method and system based on data analysis. Background Technology

[0002] With the development of enterprise informatization, the sources of financial data are becoming increasingly diversified, covering multiple independent business systems such as ERP, CRM, and supply chain management, and the scale and complexity of the data continue to increase. At the same time, enterprises are placing higher demands on the efficiency of financial data processing, the accuracy of analysis, and the automation of processes. Inefficient or erroneous data processing in business decision-making can lead to distorted analytical conclusions.

[0003] In related technologies, automated financial analysis typically relies on pre-defined data transformation rules and fixed multidimensional analysis models. When a new data source is connected or the data structure of the existing system changes, the established transformation rules may become incompatible, and the static data model may produce logical errors, leading to the failure of the automated process. Existing methods suffer from problems such as inability to automatically adapt to changes in data sources, difficulty in ensuring the consistency of data integration logic, and high system maintenance costs, affecting the continuity and reliability of the automated financial analysis process. Summary of the Invention

[0004] This application provides a financial data analysis method and system based on data analysis, which solves the problem of interruption in the automated financial analysis process in the prior art due to inconsistent standards of multi-source data and logical conflicts in data structure. It realizes fully automated processing from heterogeneous data access to analysis report generation.

[0005] This application provides a data-based financial data analysis method, which is applied to a data-based financial data analysis system, including:

[0006] Extract heterogeneous raw financial data distributed across different business systems;

[0007] The heterogeneous raw financial data is cleaned based on the preset attribute fingerprint features, and the calculation methods are standardized and aligned to obtain standard financial data.

[0008] Standard financial data are correlated based on the financial analysis topology model and mapped to a unified multidimensional data theme to obtain multidimensional financial data.

[0009] Based on the preset analytical indicator logic, logical operations are performed on multidimensional financial data to obtain key indicator data;

[0010] Establish a mapping relationship matrix between key indicator data and preset report templates, populate the key indicator data into the preset report templates and generate analysis charts, and finally output a financial analysis report.

[0011] Furthermore, the extraction of heterogeneous raw financial data distributed across different business systems includes:

[0012] Establish communication connections with the databases of various business systems to obtain metadata information of heterogeneous financial raw data stored in different business systems;

[0013] Based on metadata information, determine the data capture frequency and data structure, and extract heterogeneous raw financial data containing transaction records and account balances from different business systems.

[0014] Furthermore, the process of cleaning heterogeneous raw financial data based on preset attribute fingerprint features and standardizing and aligning calculation methods to obtain standard financial data includes:

[0015] Extract the numerical distribution characteristics and field length characteristics of heterogeneous financial raw data, and generate attribute fingerprint features;

[0016] Data cleaning is completed by identifying and removing duplicate and logically conflicting records in heterogeneous raw financial data based on attribute fingerprint features.

[0017] The cleaned data is converted according to a unified unit of measurement and time granularity to achieve standardized alignment of the calculation scope.

[0018] Furthermore, the standard financial data is correlated according to the financial analysis topology model and mapped to a unified multidimensional data theme to obtain multidimensional financial data, including:

[0019] Define multiple data dimension nodes and the logical connection relationships between nodes in the financial analysis topology model;

[0020] Standard financial data is assigned to corresponding dimension nodes, and data is aggregated according to logical connections to generate multidimensional financial data.

[0021] Furthermore, the step of performing logical operations on multidimensional financial data according to preset analytical indicator logic to obtain key indicator data includes:

[0022] Extract the preset time dimension attributes and monetary dimension attributes from multidimensional financial data;

[0023] The time dimension attribute and the monetary dimension attribute are calculated according to the mathematical operation rules defined by the preset analysis indicator logic.

[0024] Furthermore, establishing the mapping relationship matrix between key indicator data and preset report templates includes:

[0025] Scan the data in the preset report template to fill in the placeholders, and extract the identification information of the placeholders;

[0026] The key indicator data is matched with the identifier information of the placeholders to generate a mapping association matrix that records the correspondence between indicator names and placeholder positions.

[0027] Furthermore, the step of filling key indicator data into a preset report template and generating analysis charts includes:

[0028] Locate the fill position in the preset report template based on the mapping correlation matrix;

[0029] Write the key indicator data into the corresponding fill positions, and generate analysis charts in the preset report template based on the key indicator data.

[0030] Furthermore, the standardized alignment calculation caliber includes:

[0031] Obtain the calculation frequency parameters of standard financial data. If the calculation frequency parameters are inconsistent, resample the standard financial data according to the smallest time unit to ensure the consistency of the calculation caliber in the time dimension.

[0032] Furthermore, the definition of multiple data dimension nodes and the logical connections between nodes in the financial analysis topology model includes:

[0033] Detect whether there are circular references between dimension nodes. If a circular reference exists, block the circular path according to the dimension hierarchy priority.

[0034] This application provides a data-based financial data analysis system to implement a data-based financial data analysis method, including: a data extraction module, a standardization module, a data mapping module, an indicator analysis module, and a report output module;

[0035] The data extraction module is used to extract heterogeneous raw financial data distributed across different business systems.

[0036] The standardization module is used to clean heterogeneous raw financial data according to preset attribute fingerprint features, and standardize and align the calculation caliber to obtain standard financial data.

[0037] The data mapping module is used to associate standard financial data according to the financial analysis topology model and map it to a unified multidimensional data theme to obtain multidimensional financial data.

[0038] The indicator analysis module is used to perform logical operations on multidimensional financial data according to preset analysis indicator logic to obtain key indicator data.

[0039] The report output module is used to establish a mapping relationship matrix between key indicator data and preset report templates, fill the key indicator data into the preset report templates and generate analysis charts, and finally output a financial analysis report.

[0040] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0041] This application provides a financial data analysis method and system based on data analysis, which extracts heterogeneous raw financial data; cleans and standardizes the data according to attribute fingerprint features to obtain standard financial data; maps the standard financial data according to a financial analysis topology model to obtain multidimensional financial data; performs calculations on the multidimensional financial data according to the analysis indicator logic to obtain key indicator data; establishes a mapping relationship matrix between the key indicator data and the report template, fills the template and generates charts, and outputs a financial analysis report.

[0042] In this process, the method generates attribute fingerprints by extracting statistical features of the data, automates data cleaning and caliber alignment, and reduces reliance on manually predefined rules.

[0043] Furthermore, by constructing a financial analysis topology model and verifying the logical connections between dimension nodes during association mapping, structural conflicts such as circular references can be identified and blocked.

[0044] Furthermore, by dynamically binding the calculated key indicator data with the placeholders in the report template through a mapping correlation matrix, the updates to the report content and the adjustments to the template style are independent of each other, thereby enhancing the system's adaptability and maintainability. Attached Figure Description

[0045] Figure 1 A flowchart of a financial data analysis method based on data analysis is provided for embodiments of this application;

[0046] Figure 2 This is a schematic diagram of the structure of a financial data analysis system based on data analysis, provided as an embodiment of this application. Detailed Implementation

[0047] This application provides a data analysis-based financial data analysis method and system, which solves the problem of interruption in the automated financial analysis process in the prior art due to inconsistent standards of multi-source data and logical conflicts in data structure. By extracting heterogeneous raw financial data distributed in different business systems, cleaning it according to preset attribute fingerprint features and performing standardized alignment of calculation methods to obtain standard financial data, the standard financial data is associated and mapped to multi-dimensional data topics according to the financial analysis topology model to obtain multi-dimensional financial data, and logical operations are performed on the multi-dimensional financial data according to preset analysis indicator logic to obtain key indicator data. A mapping association matrix between key indicator data and preset report templates is established and filled to generate analysis charts, and finally, a financial analysis report is output, realizing the fully automated processing from heterogeneous data access to analysis report generation.

[0048] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0049] like Figure 1 As shown, this application provides a data-based financial data analysis method, which is applied to a data-based financial data analysis system and includes:

[0050] Extract heterogeneous raw financial data distributed across different business systems;

[0051] The heterogeneous raw financial data is cleaned based on the preset attribute fingerprint features, and the calculation methods are standardized and aligned to obtain standard financial data.

[0052] Standard financial data are correlated based on the financial analysis topology model and mapped to a unified multidimensional data theme to obtain multidimensional financial data.

[0053] Based on the preset analytical indicator logic, logical operations are performed on multidimensional financial data to obtain key indicator data;

[0054] Establish a mapping relationship matrix between key indicator data and preset report templates, populate the key indicator data into the preset report templates and generate analysis charts, and finally output a financial analysis report.

[0055] In this embodiment, the problem arises because the company's original financial data is scattered across multiple heterogeneous business systems, such as ERP, CRM, and OA systems, and the data patterns, formats, and calculation methods are inconsistent, resulting in the inability to efficiently, automatically, and accurately generate comprehensive financial analysis reports.

[0056] Furthermore, the extraction of heterogeneous raw financial data distributed across different business systems includes:

[0057] Establish communication connections with the databases of various business systems to obtain metadata information of heterogeneous financial raw data stored in different business systems;

[0058] Based on metadata information, determine the data capture frequency and data structure, and extract heterogeneous raw financial data containing transaction records and account balances from different business systems.

[0059] In this embodiment, connection parameters for each business system's data source need to be pre-configured. For relational databases that support JDBC, such as Oracle and MySQL, the corresponding JDBC driver is used to establish a connection; for systems that provide data via API, HTTPS protocol and authentication information (i.e., API Key) are configured to establish a network connection.

[0060] After the connection is established, query the system directory tables of each data source to obtain metadata information such as table name, field name, data type, and whether it is a primary key for the heterogeneous financial raw data.

[0061] Based on the business logic requirements of daily and monthly reports and the fields in metadata indicating data update time, a data fetching frequency is set for each data table. According to the data structure mapping rules defined in the metadata, standardized data query statements are constructed to extract specified fields from each business system.

[0062] These fields must include transaction records that form the basis of financial analysis, such as voucher number, transaction date, and debit / credit amounts, as well as static data such as account balances and customer / supplier master data, which together constitute the heterogeneous raw financial dataset to be processed.

[0063] Furthermore, the process of cleaning heterogeneous raw financial data based on preset attribute fingerprint features and standardizing and aligning calculation methods to obtain standard financial data includes:

[0064] Extract the numerical distribution characteristics and field length characteristics of heterogeneous financial raw data, and generate attribute fingerprint features;

[0065] Data cleaning is completed by identifying and removing duplicate and logically conflicting records in heterogeneous raw financial data based on attribute fingerprint features.

[0066] The cleaned data is converted according to a unified unit of measurement and time granularity to achieve standardized alignment of the calculation scope.

[0067] In this embodiment, a descriptive statistical feature vector, called attribute fingerprint feature, is calculated for each field to be cleaned.

[0068] Numeric fields, such as amounts `AMOUNT`: Calculate their mean, standard deviation, and number of significant figures.

[0069] mean : .in, This is the total number of records in this field. It is the first The value of this field for each record. This reflects the central tendency of the data. 'i' represents the record number or index, used to refer to the i-th specific data record in the dataset.

[0070] Standard deviation : . Used to measure the degree of dispersion of data.

[0071] Significant digits This is determined by analyzing the patterns of consecutive zero values ​​in the data or based on system configuration.

[0072] For text fields, such as subject name `ACCOUNT_NAME`: calculate its maximum and average length.

[0073] Maximum length : The unit is the number of characters.

[0074] average length : The unit is the number of characters.

[0075] Finally, the field Attribute fingerprint features It can be represented as: j represents the field number or index, used to refer to the j-th field in the dataset.

[0076] Data cleaning was performed based on the Laida criterion and similarity calculations, specifically including:

[0077] For fields If the value of a certain record satisfy ,in and If the value is a statistical value from historical or sample data for this field, it is considered a low-probability event, and the record is marked as an outlier and removed.

[0078] Calculate two records and Cosine similarity of fingerprint vectors across all fields. For records with multiple fields, cosine similarity of all numeric fields. and text fields After normalization, they are concatenated into a total feature vector. and Cosine similarity .like threshold If the value is typically set to 0.95 to 0.99, then two records are considered duplicates, and only one is kept.

[0079] According to predefined business rules, such as "current period depreciation" should not be greater than "original asset value", records that violate such rules are marked as logical conflict records and removed.

[0080] The cleaned data needs to be standardized and aligned to unify the units of measurement and time reference.

[0081] Unit conversions are performed according to standardized rules. For example, amounts in various currencies are standardized to the functional currency, such as the Renminbi "yuan," based on internationally accepted exchange rates.

[0082] Furthermore, the standard financial data is correlated according to the financial analysis topology model and mapped to a unified multidimensional data theme to obtain multidimensional financial data, including:

[0083] Define multiple data dimension nodes and the logical connection relationships between nodes in the financial analysis topology model;

[0084] Standard financial data is assigned to corresponding dimension nodes, and data is aggregated according to logical connections to generate multidimensional financial data.

[0085] In this embodiment, the financial analysis topology model is a directed graph. It is used to formally describe the multidimensional analysis structure of financial data.

[0086] Data dimension node set : Each element represents an analytical dimension, such as "Institution," "Product," "Accounting Period," or "Customer"; k represents the node's sequence number or index, used to refer to the k-th dimension node in the financial analysis topology model. These nodes are mapped from specific fields of standard financial data (such as `ORG_CODE`, `PRODUCT_ID`).

[0087] logical connection relationship set : It is a set of directed edges. An edge... Indicates from child node To the parent node Aggregation relationships, such as "Shanghai Branch". Belongs to the "East China Business Division" This graph structure is loaded from the configuration file during initialization.

[0088] Each record of standard financial data is assigned to the corresponding leaf dimension node in the financial analysis topology model based on its dimension field values. When a specific combination of dimensions needs to be analyzed, such as "sales revenue of each institution by product", the data is aggregated from the child nodes to the parent node according to the path defined by the logical connection relationship set, forming a multidimensional financial data cube that can be processed online.

[0089] Furthermore, the step of performing logical operations on multidimensional financial data according to preset analytical indicator logic to obtain key indicator data includes:

[0090] Extract the preset time dimension attributes and monetary dimension attributes from multidimensional financial data;

[0091] According to the mathematical operation rules defined by the preset analysis indicator logic, the time dimension attribute and the monetary dimension attribute are summed, averaged or ratio calculated.

[0092] In this embodiment, multidimensional financial data is calculated according to a preset analytical indicator logic. The analytical indicator logic is a template that encapsulates mathematical formulas and data retrieval rules.

[0093] For example, first, from multidimensional financial data, locate the beginning and ending balances of the "Accounts Receivable" account (amount-based attribute) and the cumulative amount of "Main Business Revenue" during the calculation period. Then, calculate the accounts receivable turnover days using the formula to obtain the key indicator data.

[0094] Accounts receivable turnover days The calculation formula is as follows: ; It is used to measure the average number of days it takes for a company to collect its accounts receivable.

[0095] in, This is the average accounts receivable balance for the calculation period. It is typically calculated using the beginning balance and the period balance. This represents the total amount of credit sales revenue during the calculation period.

[0096] The number of days in the calculation period, such as 360 days for a year; The calculated accounts receivable turnover days.

[0097] Furthermore, establishing the mapping relationship matrix between key indicator data and preset report templates includes:

[0098] Scan the data in the preset report template to fill in the placeholders, and extract the identification information of the placeholders;

[0099] The key indicator data is matched with the identifier information of the placeholders to generate a mapping association matrix that records the correspondence between indicator names and placeholder positions.

[0100] In this embodiment, the preset report template is a structured document containing data placeholders. The template is scanned, and all placeholders and their location information are parsed out, such as bookmarks in a Word document or specific cell addresses in an Excel file. Their identification information, i.e., placeholder name strings, is extracted.

[0101] The mapping association matrix is ​​essentially a lookup table. It matches the name of each calculated key indicator data point with all identifiers extracted from the template, including exact matches and synonym matches (the synonym table needs to be pre-configured). Upon successful matching, a record is generated: Key Indicator Data ID, Indicator Value, and Placeholder Position. The mapping association matrix reflects the one-to-one correspondence between data results and their presentation positions in the report.

[0102] Furthermore, the step of filling key indicator data into a preset report template and generating analysis charts includes:

[0103] Locate the fill position in the preset report template based on the mapping correlation matrix;

[0104] Write the key indicator data into the corresponding fill positions, and generate analysis charts in the preset report template based on the key indicator data. The analysis charts include trend charts or percentage charts.

[0105] In this embodiment, based on the records in the mapping correlation matrix, the value of each key indicator data is filled into the corresponding placeholder position in the preset report template.

[0106] For chart generation, the preset report template defines the chart type and data range. For example, when generating a management expense trend chart, the management expense indicator values ​​for each month of the past 12 months are found from the matrix and arranged into a data sequence in chronological order. Subsequently, the chart rendering engine is invoked to execute standardized plotting instructions, transforming the data sequence into charts and outputting a final financial analysis report embedding the charts and data.

[0107] Furthermore, the standardized alignment calculation caliber includes:

[0108] Obtain the calculation frequency parameters of standard financial data. If the calculation frequency parameters are inconsistent, resample the standard financial data according to the smallest time unit to ensure the consistency of the calculation caliber in the time dimension.

[0109] In this embodiment, the time granularity parameter of the data is examined, such as daily, weekly, or monthly settlement. If it is necessary to unify it to a "daily" granularity, then resampling processing is performed on the coarse-grained data. For example, weekly sales figures... Distributing the values ​​evenly across the days of the week yields the daily sequence values: , This is the number of days in the week, usually 7. After the above processing, standardized financial data with consistent definitions is obtained.

[0110] Furthermore, the definition of multiple data dimension nodes and the logical connections between nodes in the financial analysis topology model includes:

[0111] Detect whether there are circular references between dimension nodes. If a circular reference exists, block the circular path according to the dimension hierarchy priority.

[0112] In this embodiment, when defining logical connections, it is necessary to prevent circular references, such as costs allocated from department A to department B, and then costs allocated from department B to department A. A depth-first search algorithm from graph theory is used to detect whether cycles exist in the financial analysis topology model.

[0113] If a cycle is detected, the node with the lowest priority in the cycle is found according to the preset priority rules of the dimension hierarchy that reflects the enterprise's management level, such as "legal entity > profit center > cost center". An edge originating from that node is temporarily removed, thereby converting the financial analysis topology model into a directed acyclic graph. This process is called blocking the cycle path.

[0114] like Figure 2 As shown in the figure, this application provides a data analysis-based financial data analysis system to implement a data analysis-based financial data analysis method, including: a data extraction module, a standardization module, a data mapping module, an indicator analysis module, and a report output module;

[0115] The data extraction module is used to extract heterogeneous raw financial data distributed across different business systems.

[0116] The standardization module is used to clean heterogeneous raw financial data according to preset attribute fingerprint features, and standardize and align the calculation caliber to obtain standard financial data.

[0117] The data mapping module is used to associate standard financial data according to the financial analysis topology model and map it to a unified multidimensional data theme to obtain multidimensional financial data.

[0118] The indicator analysis module is used to perform logical operations on multidimensional financial data according to preset analysis indicator logic to obtain key indicator data.

[0119] The report output module is used to establish a mapping relationship matrix between key indicator data and preset report templates, fill the key indicator data into the preset report templates and generate analysis charts, and finally output a financial analysis report.

[0120] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0121] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0122] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0123] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0124] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0125] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A financial data analysis method based on data analysis, characterized in that, Includes the following steps: Extract heterogeneous raw financial data distributed across different business systems; The heterogeneous raw financial data is cleaned based on the preset attribute fingerprint features, and the calculation methods are standardized and aligned to obtain standard financial data. Standard financial data are correlated based on the financial analysis topology model and mapped to a unified multidimensional data theme to obtain multidimensional financial data. Based on the preset analytical indicator logic, logical operations are performed on multidimensional financial data to obtain key indicator data; Establish a mapping relationship matrix between key indicator data and preset report templates, populate the key indicator data into the preset report templates and generate analysis charts, and finally output a financial analysis report.

2. The financial data analysis method based on data analysis as described in claim 1, characterized in that, The extraction of heterogeneous raw financial data distributed across different business systems includes: Establish communication connections with the databases of various business systems to obtain metadata information of heterogeneous financial raw data stored in different business systems; Based on metadata information, determine the data capture frequency and data structure, and extract heterogeneous raw financial data containing transaction records and account balances from different business systems.

3. The financial data analysis method based on data analysis as described in claim 1, characterized in that, The process of cleaning heterogeneous raw financial data based on preset attribute fingerprint features and standardizing and aligning calculation methods to obtain standard financial data includes: Extract the numerical distribution characteristics and field length characteristics of heterogeneous financial raw data, and generate attribute fingerprint features; Data cleaning is completed by identifying and removing duplicate and logically conflicting records in heterogeneous raw financial data based on attribute fingerprint features. The cleaned data is converted according to a unified unit of measurement and time granularity to achieve standardized alignment of the calculation scope.

4. The financial data analysis method based on data analysis as described in claim 1, characterized in that, The process involves associating standard financial data according to a financial analysis topology model and mapping it to a unified multidimensional data theme to obtain multidimensional financial data, including: Define multiple data dimension nodes and the logical connection relationships between nodes in the financial analysis topology model; Standard financial data is assigned to corresponding dimension nodes, and data is aggregated according to logical connections to generate multidimensional financial data.

5. The financial data analysis method based on data analysis as described in claim 1, characterized in that, The process of performing logical operations on multidimensional financial data based on preset analytical indicator logic to obtain key indicator data includes: Extract the preset time dimension attributes and monetary dimension attributes from multidimensional financial data; The time dimension attribute and the monetary dimension attribute are calculated according to the mathematical operation rules defined by the preset analysis indicator logic.

6. The financial data analysis method based on data analysis as described in claim 1, characterized in that, The establishment of a mapping relationship matrix between key indicator data and preset report templates includes: Scan the data in the preset report template to fill in the placeholders, and extract the identification information of the placeholders; The key indicator data is matched with the identifier information of the placeholders to generate a mapping association matrix that records the correspondence between indicator names and placeholder positions.

7. The financial data analysis method based on data analysis as described in claim 1, characterized in that, The step of filling key indicator data into a preset report template and generating analysis charts includes: Locate the fill position in the preset report template based on the mapping correlation matrix; Write the key indicator data into the corresponding fill positions, and generate analysis charts in the preset report template based on the key indicator data.

8. The financial data analysis method based on data analysis as described in claim 3, characterized in that, The standardized alignment calculation caliber includes: Obtain the calculation frequency parameters of standard financial data. If the calculation frequency parameters are inconsistent, resample the standard financial data according to the smallest time unit to ensure the consistency of the calculation caliber in the time dimension.

9. The financial data analysis method based on data analysis as described in claim 4, characterized in that, The definition of multiple data dimension nodes and the logical connections between nodes in the financial analysis topology model includes: Detect whether there are circular references between dimension nodes. If a circular reference exists, block the circular path according to the dimension hierarchy priority.

10. A data-based financial data analysis system, used to implement the data-based financial data analysis method according to any one of claims 1-9, characterized in that, include: Data extraction module, standardization module, data mapping module, indicator analysis module, report output module; The data extraction module is used to extract heterogeneous raw financial data distributed across different business systems. The standardization module is used to clean heterogeneous raw financial data according to preset attribute fingerprint features, and standardize and align the calculation caliber to obtain standard financial data. The data mapping module is used to associate standard financial data according to the financial analysis topology model and map it to a unified multidimensional data theme to obtain multidimensional financial data. The indicator analysis module is used to perform logical operations on multidimensional financial data according to preset analysis indicator logic to obtain key indicator data. The report output module is used to establish a mapping relationship matrix between key indicator data and preset report templates, fill the key indicator data into the preset report templates and generate analysis charts, and finally output a financial analysis report.

Images