Intelligent financial analysis method and apparatus based on automated bookkeeping big data

Abstract

The present invention relates to the field of intelligent financial analysis methods. Provided are an intelligent financial analysis method and apparatus based on automated bookkeeping big data. The intelligent financial analysis apparatus based on automated bookkeeping big data comprises: a data collection module, a dynamic data monitoring module, a data pre-processing module, a data integrity check module, a financial model construction module, a dynamic model adjustment module, an intelligent analysis module, a multi-dimensional analysis result verification module, a result display module, a user feedback module and a system self-learning module. In the method, by means of dynamic data monitoring and the provision of a backup data source interface, when a data source changes or an interface fails, it can automatically switch to a backup data source or take a remedial action, so as to ensure the continuous and stable acquisition of data.

Description

Financial intelligent analysis method and device based on automatic accounting big data TECHNICAL FIELD

[0001] The present application relates to the technical field of financial intelligent analysis method, in particular to a financial intelligent analysis method and device based on automatic accounting big data. BACKGROUND

[0002] The financial intelligent analysis method and device based on automatic accounting big data mainly comprises a data collection module, a data preprocessing module, a financial model construction module, an intelligent analysis module and a result display module. First, the data collection module automatically extracts data from various financial systems, bank accounts, electronic invoices and transaction platforms, covering financial information such as income, expenditure, assets and liabilities. Then, the data preprocessing module cleans, classifies and integrates these data, removes duplicate and abnormal data, and ensures the accuracy and consistency of the data. Next, the financial model construction module constructs financial models based on historical data and industry standards, which can include budget prediction models, cash flow analysis models and cost control models. The intelligent analysis module uses machine learning algorithms and big data analysis techniques to train and optimize the models, so as to automatically identify financial anomalies, predict financial risks, and generate corresponding decision recommendations. Finally, the result display module presents the analysis results to users in the form of charts, reports and visual dashboards, helping enterprise managers make more intelligent financial decisions.

[0003] Although this system can significantly improve the efficiency and accuracy of financial analysis, there are still some deficiencies. Since the data collection module relies on the data interface of external systems, once the data source changes or the interface fails, it may cause instability and delay in data acquisition, thereby affecting the real-time performance of the overall analysis. The financial model construction and intelligent analysis module mainly rely on historical data and algorithm training, and if the historical data is biased or incomplete, it may lead to inaccurate prediction results of the model.

[0004] SUMMARY

[0005] (I) Technical problems solved

[0006] To solve the problems of the prior art, the present application provides a financial intelligent analysis method and device based on automatic accounting big data, which solves the problems of instability and delay in data acquisition caused by changes in data sources or interface failures, thereby affecting the real-time performance of the overall analysis. The financial model construction and intelligent analysis module mainly rely on historical data and algorithm training, and if the historical data is biased or incomplete, it may lead to inaccurate prediction results of the model.

[0007] (II) Technical solutions

[0008] To achieve the above object, the present application is realized by the following technical solutions:

[0009] A financial intelligent analysis device based on automatic accounting big data, comprising: a data collection module, a data dynamic monitoring module, a data preprocessing module, a data integrity verification module, a financial model construction module, a model dynamic adjustment module, an intelligent analysis module, an analysis result multi-dimensional verification module, a result display module, a user feedback module, and a system self-learning module.

[0010] A financial intelligent analysis method based on automatic accounting big data, comprising:

[0011] a. Data collection: automatically collecting financial information including income, expenditure, assets, and liabilities through multiple data source interfaces;

[0012] b. Data dynamic monitoring: monitoring the status of data source interfaces in real time, and automatically switching to backup data sources or taking remedial measures when data sources change or interfaces fail, to ensure continuous and stable data acquisition;

[0013] c. Data preprocessing: cleaning, classifying, and integrating collected data to remove duplicate and abnormal data, wherein the data cleaning step includes calculating the mean μ and standard deviation σ of the data, and performing standardization processing on the data through the formula ;

[0014] d. Data integrity verification: introducing a data integrity verification mechanism, using multi-source data cross verification to ensure data accuracy and consistency, and using a consistency verification formula ; i and Y i are corresponding data items from different data sources;

[0015] e. Data completion: using machine learning algorithms to intelligently complete missing or incomplete historical data, specifically using the K-nearest neighbor algorithm, whose formula is ; i where y

[0016] f. Financial model construction: based on the completed and verified data, constructing financial models including budget prediction, cash flow analysis, and cost control, and the prediction of the model uses a linear regression model in the form of Y = β0 + β1X1 + … + β n X n ;

[0017] g. Model dynamic adjustment: automatically adjusting and optimizing the financial model according to newly collected data and real-time market dynamics, using a rolling window technique, and the formula is ; where α is the smoothing coefficient;

[0018] h. Intelligent analysis: Utilize the optimized financial model to perform anomaly detection, risk prediction, and decision-making suggestion generation. Anomaly detection uses an algorithm based on probability density function, which is in the form of

[0019] i. Multi-dimensional verification of analysis results: Perform multi-dimensional verification of intelligent analysis results using historical data backtesting method, formula is Where R i is the actual result, P i is the predicted result;

[0020] j. Result display: Display the analysis results to the user in the form of charts, reports, and visual dashboards;

[0021] k. User feedback mechanism: Users can provide feedback on the analysis results, and the system will further optimize the model and algorithm based on the feedback;

[0022] I. System self-learning: The system learns and optimizes itself based on user feedback and the actual application effect of the analysis results, continuously improving the accuracy and real-time performance of the overall analysis.

[0023] Preferably, the data collection module includes at least two different sources of backup data interfaces, which automatically switch to the backup interface when the main data source interface fails.

[0024] Preferably, the data dynamic monitoring step includes recording the response time and success rate of the data source interface in real time, and if an anomaly is detected, an alarm is immediately issued and the switching mechanism is automatically started.

[0025] Preferably, the data preprocessing step further includes data format standardization processing to ensure that data from different sources can be integrated and analyzed under the same standard, and the data completion algorithm is based on machine learning technology, using similar data patterns to infer and complete missing historical data.

[0026] Preferably, the data integrity verification compares the trends and patterns of historical data to automatically mark abnormal or inconsistent data records, and uses the Euclidean distance formula to calculate the difference between data points.

[0027] Preferably, the financial model construction step includes multi-dimensional modeling based on industry standards and historical data to meet the individual needs of different enterprises, and the model dynamically adjusts real-time data analysis based on a rolling time window to ensure that the model can flexibly adjust to market changes. The intelligent analysis step uses deep learning technology to identify complex financial anomalies and potential risks, using a long short-term memory network, and the model formula is h t= sigma(W h h t-1 + W x X t + b), where h t is the hidden layer state, X t is the input.

[0028] Preferably, the analysis result multi-dimensional verification includes backtesting through historical data and comparative analysis with the same industry to ensure the accuracy of the prediction results, and the result display module includes user-defined visualization options, and the user can select different display modes according to needs.

[0029] Preferably, the user feedback mechanism is based on an interactive platform, and the user can submit opinions and suggestions on the analysis results in real time through the platform, and the system self-learning function continuously adjusts and optimizes the model parameters and algorithms using the feedback data and the actual effect of the analysis results.

[0030] (III) Beneficial effects

[0031] The application provides a financial intelligent analysis method and device based on automatic accounting big data. The following beneficial effects are provided:

[0032] The method can automatically switch to a backup data source or take remedial measures when the data source changes or the interface fails, ensuring continuous and stable data acquisition, through dynamic data monitoring and backup data source interface setting. This effectively solves the problem of data acquisition instability and delay caused by data source problems in traditional systems, significantly improving the real-time and reliability of financial data analysis. Through data cleaning, data integrity verification, and multi-source data cross-validation, the method ensures the accuracy and consistency of financial data. Standardization processing, Euclidean distance and other mathematical methods in data processing make data preprocessing more standardized and accurate, and the data completion algorithm uses K-nearest neighbor and other machine learning techniques to intelligently complete missing or incomplete historical data, further improving the integrity of the data and the reliability of the analysis.

[0033] The application shows high flexibility and adaptability in the construction and application of financial models. The financial model construction is based on multi-dimensional historical data and industry standards, which can adapt to the individual needs of different enterprises, and through rolling window technology and dynamic adjustment mechanism, the model is automatically optimized and adjusted in real time, ensuring the accuracy and timeliness of the prediction results. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 is a flowchart of the application. DETAILED DESCRIPTION

[0035] The technical solutions in the embodiments of the present application are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0036] As shown in FIG. 1, the present application provides a financial intelligent analysis method based on automatic accounting big data, including: a. Data collection: automatically collecting financial information including income, expenditure, assets and liabilities through multiple data source interfaces, the data collection module includes at least two different source backup data interfaces, when the main data source interface fails, automatically switch to the backup interface.

[0037] b. Data dynamic monitoring: real-time monitoring of the state of the data source interface, and automatically switching to the backup data source or taking remedial measures when the data source changes or the interface fails, to ensure continuous and stable data acquisition, the data dynamic monitoring step includes real-time recording of the response time and success rate of the data source interface, if an anomaly is detected, an alarm is immediately issued and the switching mechanism is automatically started.

[0038] c. Data preprocessing: cleaning, classifying and integrating the collected data, removing duplicate data and abnormal data, wherein the data cleaning step includes calculating the mean μ and standard deviation σ of the data, and normalizing the data by the formula The data preprocessing step further includes data format standardization processing to ensure that data from different sources can be integrated and analyzed under the same standard, the data completion algorithm is based on machine learning technology, using the pattern of similar data to infer and complete the missing historical data.

[0039] d. Data integrity verification: introducing a data integrity verification mechanism, using multi-source data cross verification to ensure data accuracy and consistency, using a consistency verification formula Wherein X i and Y i are corresponding data items from different data sources, the data integrity verification automatically marks abnormal or inconsistent data records by comparing the trend and pattern of historical data, using the Euclidean distance formula To calculate the difference between data points.

[0040] e. Data completion: using machine learning algorithms to intelligently complete missing or incomplete historical data, specifically using the K-nearest neighbor algorithm, the formula is Where y i is the k nearest known data points.

[0041] f. Financial Model Construction: Based on the completed and verified data, construct a financial model including budget forecasts, cash flow analysis, and cost control, the prediction of the model uses linear regression model, the form is Y = β0 + β1X1 + … + β n X n , the financial model construction step includes multi-dimensional modeling based on industry standards and historical data to adapt to the personalized needs of different enterprises, the model dynamically adjusts real-time data analysis based on rolling time window to ensure that the model can flexibly adjust with market changes, the intelligent analysis step uses deep learning technology to identify complex financial anomalies and potential risks, using long short-term memory network, the model formula is h t = σ (W h h t-1 + W x X t + b), where h t is the hidden layer state, X t is the input.

[0042] g. Model Dynamic Adjustment: Automatically adjust and optimize the financial model according to newly collected data and real-time market dynamics, using rolling window technology, the formula is where α is the smoothing coefficient.

[0043] h. Intelligent Analysis: Use the optimized financial model to perform anomaly detection, risk prediction, and decision suggestion generation, anomaly detection uses an algorithm based on probability density function, the form is

[0044] i. Multi-dimensional verification of analysis results: Use historical data backtesting method, the formula is where R i is the actual result, P i is the predicted result, the multi-dimensional verification of analysis results includes historical data backtesting and comparison analysis with the same industry to ensure the accuracy of the predicted results, the result display module includes user-defined visualization options, users can choose different display modes according to their needs.

[0045] j. Result display: Display the analysis results to the user in the form of charts, reports, and visual dashboards.

[0046] k. User feedback mechanism: Users can provide feedback on the analysis results, the system further optimizes the model and algorithm based on the feedback, the user feedback mechanism is based on an interactive platform, users can submit their opinions and suggestions on the analysis results through the platform in real time, the system self-learning function uses feedback data and actual effects of analysis results to continuously adjust and optimize model parameters and algorithms.

[0047] I. System self-learning: The system learns and optimizes itself according to user feedback and the actual application effect of the analysis results, continuously improving the accuracy and real-time performance of the overall analysis.

[0048] A financial intelligent analysis device based on automatic accounting big data, comprising: a data collection module, a data dynamic monitoring module, a data preprocessing module, a data integrity verification module, a financial model construction module, a model dynamic adjustment module, an intelligent analysis module, an analysis result multi-dimensional verification module, a result display module, a user feedback module, and a system self-learning module.

[0049] Table 1: Data comparison table

[0050] Although embodiments of the present application have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made thereto without departing from the principles and spirit of the application, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A financial intelligent analysis device based on automatic accounting big data, characterized in that: The system comprises a data collection module, a data dynamic monitoring module, a data preprocessing module, a data integrity verification module, a financial model construction module, a model dynamic adjustment module, an intelligent analysis module, an analysis result multi-dimensional verification module, a result display module, a user feedback module, and a system self-learning module.

2. A financial intelligent analysis method based on automatic accounting big data, characterized in that, The system comprises: a. Data collection: automatically collecting financial information including income, expenditure, assets, and liabilities through multiple data source interfaces; b. Data dynamic monitoring: real-time monitoring of data source interfaces and automatically switching to backup data sources or taking remedial measures when data sources change or interfaces fail; c. Data preprocessing: The collected data is cleaned, classified and integrated, and repeated data and abnormal data are removed, wherein the data cleaning step includes calculating the mean μ and standard deviation σ of the data, and removing the data that does not satisfy the formula Standardizing the data; d. Data integrity check: Introduce data integrity check mechanism, use multi-source data cross verification to ensure data accuracy and consistency, and use consistency verification formula wherein X i and Y i are corresponding data items of different data sources; e. Data completion: For missing or incomplete historical data, intelligent completion is performed using a machine learning algorithm. Specifically, a K-Nearest Neighbor algorithm is used, and the formula is where y i are the nearest k known data points; f. Financial model building: Based on the completed and verified data, build a financial model including budget forecast, cash flow analysis and cost control, etc. The prediction of the model uses a linear regression model, which is in the form of Y = β0+ β1X1+ … + β n X n ; g. Model dynamic adjustment: automatically adjust and optimize the financial model according to newly collected data and real-time market dynamics, using rolling window techniques, formula is Wherein α is the smoothing coefficient; j. Result display: displaying analysis results to users in the form of charts, reports, and visual dashboards; h. Intelligent analysis: Using the optimized financial model, anomaly detection, risk prediction, and decision suggestion generation are performed. Anomaly detection uses an algorithm based on the probability density function, which is in the form of i. Analysis result multi-dimensional verification: the intelligent analysis result is verified in multiple dimensions, and a historical data backtest method is adopted, and the formula is where R i is the actual result, P i is the predicted result; k. User feedback mechanism: users can provide feedback on analysis results, and the system further optimizes models and algorithms based on feedback; I. System self-learning: the system learns and optimizes itself based on user feedback and the actual application effect of analysis results, continuously improving the accuracy and real-time performance of overall analysis. The data collection module includes at least two backup data interfaces from different sources, which automatically switch to backup interfaces when the main data source interface fails. 3.The financial intelligent analysis method based on automatic accounting big data according to claim 2, characterized in that: The data dynamic monitoring step includes real-time recording of data source interface response time and success rate, and if an anomaly is detected, an alarm is immediately issued and the switching mechanism is automatically started.

4. The financial intelligence analysis method based on automatic accounting big data according to claim 2, characterized in that: The data preprocessing step further includes data format standardization processing to ensure that data from different sources can be integrated and analyzed under the same standard, and the data completion algorithm uses similar data patterns to infer and complete missing historical data based on machine learning technology.

5. The financial intelligence analysis method based on automatic accounting big data according to claim 2, characterized in that: The analysis result multi-dimensional verification includes historical data backtesting and industry comparison analysis to ensure the accuracy of the prediction results, and the result display module includes user-defined visualization options, allowing users to choose different display methods according to their needs.

6. The financial intelligence analysis method based on automatic accounting big data according to claim 2, characterized in that: The data integrity check automatically flags anomalous or inconsistent data records by comparing trends and patterns in historical data, using the Euclidean distance formula Calculate the difference between data points.

7. The financial intelligence analysis method based on automatic accounting big data according to claim 2, characterized in that: The financial model construction step includes multi-dimensional modeling based on industry standards and historical data to adapt to the personalized needs of different enterprises, the model dynamically adjusts real-time data analysis based on a rolling time window, ensuring that the model can be flexibly adjusted as the market changes, and the intelligent analysis step uses deep learning technology to identify complex financial anomalies and potential risks, using a long short-term memory network, the model formula is h t = σ(W h h t-1 +W x X t +b), where h t is the hidden layer state, X t is the input.

8. The financial intelligence analysis method based on automatic accounting big data according to claim 2, characterized in that: The user feedback mechanism is based on an interactive platform, allowing users to submit opinions and suggestions on analysis results in real time, and the system self-learning function uses feedback data and the actual effect of analysis results to continuously adjust and optimize model parameters and algorithms. 9.The financial intelligent analysis method based on automatic accounting big data according to claim 2, characterized in that: ​

Images