Search

CN-121998782-A - Automatic analysis system and method for enterprise financial report

CN121998782ACN 121998782 ACN121998782 ACN 121998782ACN-121998782-A

Abstract

The invention discloses an automatic analysis system and method for an enterprise financial statement, which relate to the technical field of financial analysis and comprise an enterprise and data unit, an analysis processing unit and a result processing unit; the enterprise and the data unit divide and classify types of the enterprise, extract and collect data of the enterprise after the enterprise is divided, analyze the data after the data is collected, build an enterprise selection model according to different enterprises through a large model technology, build a corresponding financial statement analysis scheme according to the classified types of the enterprise and analyze the enterprise, and the result processing unit performs risk early warning according to the analysis processed result after the analysis processing. After the analysis processing is finished, the risk early warning processing is carried out on the generated result, grading is carried out during the risk early warning processing, a corresponding processing scheme is formulated, and the accuracy during risk early warning and the efficiency and accuracy of risk processing during risk early warning are improved.

Inventors

  • CHEN WENJING

Assignees

  • 浙江金融职业学院

Dates

Publication Date
20260508
Application Date
20251216

Claims (10)

  1. 1. An automatic analysis system for an enterprise financial statement is characterized by comprising an enterprise and data unit, an analysis processing unit and a result processing unit; The enterprise and the data unit divide and classify the types of the enterprise, extract and collect the data of the enterprise after the enterprise is divided, analyze the data after the collection is completed, and adjust the classified enterprise after the analysis; The analysis processing unit builds an enterprise selection model according to different enterprises through a large model technology, and builds a corresponding financial statement analysis scheme for analysis according to classified enterprise types; And the result processing unit performs risk early warning according to the analysis processed result after the analysis processing, and processes the risk after the risk early warning processing is completed.
  2. 2. The system for automatically analyzing the financial statement of the enterprise as set forth in claim 1, wherein the enterprise and data unit comprises an enterprise classification module, a data extraction module and a data analysis module, wherein the enterprise classification module classifies the enterprise in the following manner; (1) Classification according to legal forms, including individual business, partnership, finite liability company, stock finite company, national enterprises and foreign enterprises; (2) Classification by industry attribute including agriculture, forestry, animal husbandry, fishery, mining, manufacturing, electricity, heat, gas and water production, supply, construction, wholesale and retail industries, transportation, storage and postal industries, accommodation and catering industries, information transmission, software and information technology service industries, financial industries, residential industries, business service industries, and scientific research and technical service industries; (3) Classification by enterprise scale, including industry, retail, software and information technology services, and construction; (4) Classifying according to all manufacturing properties, including national enterprises, collective enterprises, civil enterprises, foreign enterprises and mixed all manufacturing enterprises; (5) The system is classified according to the region range and the financing structure, wherein the enterprise registration place comprises domestic enterprises/cross-country enterprises, eastern enterprises/western enterprises, first-line urban/third-line and fourth-line urban enterprises, the operation coverage area comprises local enterprises, regional enterprises, national enterprises and cross-country enterprises, and the financing structure comprises a stock right financing type, a credited financing type and self-raising financing type.
  3. 3. The system for automatically analyzing the financial statement of the enterprise according to claim 2, wherein the data extraction module extracts data in real time through a web crawler, an API interface and the like, a data extraction model is formulated during enterprise feature extraction, and the data extraction model comprises basic surface features, operation and financial features, innovation and strength features, risk and honesty features, public opinion features, association and network features and strategy and behavior features; (1) Basic surface features include established years, registered capital, status of market, industry and region to which they belong; (2) The business and financial features include revenue, profit, liability, cash flow, and employee count; (3) Innovation and strength characteristics comprise the number/type of patents, the number of trademarks, the research and development investment ratio and the qualification of high and new technology enterprises; (4) The risk and integrity features include the number of legal litigation, the number of executed persons, administrative penalty records and equity freezing information; (5) Public opinion features include news exposure, public opinion emotion score, bidding liveness and market share; (6) Association with network characteristics including stockholder chain depth, number of subsidiaries, supply chain partners and competitors; (7) The strategic and behavioral characteristics include business scope keywords, strategic collaboration events, investment and purchase events, and overseas expansion dynamics; In the process of extracting enterprise data features, cleaning, de-duplication and formatting the extracted feature data, associating and merging the feature data of the same enterprise from different sources, performing multi-source data cross-validation processing on missing and inconsistent data, building a data feature storage library, and formulating a periodic update scheme, wherein the update scheme comprises a first update scheme, a second update scheme and a third update scheme, the first update scheme is update adjustment within one week, the second update scheme is update adjustment within half month, and the third update scheme is update adjustment within one month.
  4. 4. The system for automatically analyzing financial statements of enterprises of claim 3, wherein the data analysis module is configured to analyze the extracted characteristic data and to collect and analyze classified enterprises during the analysis of the enterprises, wherein the analyzed enterprise types comprise policy-affected enterprises, market-affected enterprises and double-affected enterprises, wherein the policy-affected enterprises are mainly affected by policies, wherein the market-affected enterprises are mainly affected by market competition and large environmental impact, and wherein the double-affected enterprises are mainly affected by policies and market environments and policies; And the data analysis scheme includes a first analysis scheme, a second analysis scheme, a third analysis scheme, and an abnormal data analysis scheme, wherein the first analysis scheme, the second analysis scheme, the third analysis scheme, and the abnormal data analysis scheme include a financial data feature analysis, a supply chain data feature analysis, a customer numerical control feature analysis, and an industrial production data feature analysis when performing analysis, wherein the first analysis scheme has a maximum number of features analyzed when analyzing, and a combined business segment and an on-line expert population, wherein the second analysis scheme has a medium number of features analyzed when analyzing, and a combined business segment and an on-line expert population, wherein the third analysis scheme has a minimum number of features analyzed when analyzing, and a combined business segment and an on-line expert population, wherein the abnormal data analysis scheme performs individual analysis on abnormal data generated when analyzing, and the abnormal data includes numerical abnormality, timing abnormality, and associated abnormality, performs combined processing on the generated problems using expert and on-line personnel in the industry when analyzing, collecting the generated problems, and after completion of the collection, making an optimal adjustment scheme, and performing optimal adjustment on the data analysis scheme.
  5. 5. The system of claim 4, wherein the analysis processing unit comprises a selection analysis module, an analysis supervision module and a period processing module, wherein the selection analysis module establishes a policy-affected enterprise analysis processing scheme, a market-affected enterprise analysis processing scheme and a double-affected enterprise analysis processing scheme through a large model technology, when the financial report of an enterprise is analyzed, the key characteristics of the enterprise are input into the large model, the corresponding analysis processing scheme can be automatically selected, and the enterprise can also manually input the characteristics, so that the accuracy of the analysis processing is improved; In the analysis processing process of the policy-influencing enterprise, the policy issued by the official of the region is referred, the policy files issued by the official website and the public number are acquired in real time, and the video account number and professor expert for policy interpretation are subjected to association analysis; The market-influencing enterprise analysis processing scheme is used for carrying out contrast analysis on industrial data and macroscopic economic indexes, adopting a constant exchange rate restated report or analyzing exchange rate sensitivity, analyzing fluctuation of macroscopic economic cycle, including economic expansion period and economic decay period, analyzing and processing market supply and demand ends and competition patterns, and simultaneously analyzing capital market emotion and industrial technology reform and industrial upgrading; The analysis processing scheme of the dual-influence enterprise simultaneously carries out dual analysis on policies and market environments, a report standardization template is arranged in the analysis process, the standardization template supports T+0 to generate an asset liability table, a profit table, a market trend table and the like in real time, a credential is automatically generated through an AI algorithm, a comprehensive evaluation model is constructed for comprehensive evaluation by combining multidimensional indexes such as DuPont analysis and a Vol scoring method, and finally an analysis evaluation template is generated according to the standardization template; And respectively constructing a primary analysis scheme, a secondary analysis scheme and a tertiary analysis scheme in the policy-affected enterprise analysis and treatment scheme and the market-affected enterprise analysis and treatment scheme, wherein the primary analysis scheme performs financial analysis on a large-sized head enterprise, the secondary analysis scheme performs financial analysis on a medium-sized large-sized enterprise, the tertiary analysis scheme performs financial analysis on a medium-sized and small-sized enterprise, and a data storage library is constructed after analysis is completed to store result data generated during analysis and perform identification treatment on the stored data.
  6. 6. The system of claim 5, wherein the analysis supervision module formulates a comprehensive supervision scheme to supervise analysis results and processes of the financial statement, the comprehensive supervision scheme comprises a first supervision scheme, a second supervision scheme and a third supervision scheme, the first supervision scheme carries out repeated disc detection analysis on the analyzed financial statement by an expert within twelve hours after the financial statement is analyzed, the second supervision scheme carries out repeated disc detection analysis on the analyzed financial statement by the expert within twenty-four hours after the financial statement is analyzed, the third supervision scheme carries out repeated disc detection analysis on the analyzed financial statement by the expert within forty-eight hours after the financial statement is analyzed, a detection analysis database is built to store the analyzed results, and the stored data are identified and classified when the stored results are stored; And analyzing the results after detection and analysis in the first supervision scheme, the second supervision scheme and the third supervision scheme respectively, wherein the results after analysis are the first level, the second level and the third level, the results after analysis are the worst, the results after analysis are the second level and the middle, the results after analysis are the third level, the time for carrying out the double-disc detection and analysis is shortened by fifty percent when the results after analysis are the first level, the time for carrying out the double-disc detection and analysis is shortened by seventy percent when the results after analysis are the second level, the time for carrying out the double-disc detection and analysis is shortened by ninety percent when the results after analysis are the third level, the time for carrying out the double-disc analysis is shortened again when the results after the double-disc analysis are abnormal, and the time is sequentially shortened as long as the results after the double-disc analysis are abnormal.
  7. 7. The system for automatically analyzing the financial statement of the enterprise as set forth in claim 6, wherein the period processing module formulates a period analysis scheme, performs multi-angle investigation and analysis on market environment and policies in a specified period, determines core indexes which need to be paid attention to regularly according to industries in which the enterprise is located, formulates a dynamic analysis method, and periodically applies a PEST framework to perform arrangement analysis on the latest data and events collected and evaluate the specific influence of the latest data and events on the enterprise; Meanwhile, the enterprise scale is divided, when the divided enterprise scale is large, the comprehensive core index in the industry can be tracked, when the divided enterprise scale is small and medium, three to five core indexes related to the industry can be tracked, future qualitative judgment and expected data are collected for enterprises and consumers, and after the periodic investigation analysis is completed, readjustment treatment is carried out on the divided enterprise scale, enterprise type and enterprise characteristics.
  8. 8. The system of claim 1, wherein the result processing unit comprises a risk early warning module and a risk processing module, the risk early warning module formulates a total risk early warning scheme to perform risk early warning processing on the enterprise, the total risk early warning scheme comprises a first risk early warning scheme, a second risk early warning scheme and a third risk early warning scheme, the first risk early warning scheme has the highest risk early warning level, the second risk early warning scheme has the medium risk early warning level, the third risk early warning scheme has the lowest risk early warning level, and the issued policy file and the result of the large-environment market investigation are referred to when the risk level classification processing is performed.
  9. 9. The risk processing module formulates a risk processing scheme when a risk is generated, and the risk processing scheme comprises a first processing scheme, a second processing scheme and a third processing scheme, wherein the first processing scheme processes the problem generated by the first risk early warning scheme, the second processing scheme processes the problem generated by the second risk early warning scheme, the third processing scheme processes the problem generated by the third processing scheme, the first processing scheme processes and adjusts at a first time, the second processing scheme can postpone a point to process and adjust, and the third processing scheme can process and adjust at a later point.
  10. 10. An automatic analysis method for enterprise financial statements is characterized in that the automatic analysis method for enterprise financial statements is applicable to the automatic analysis system for enterprise financial statements as set forth in claim 8, and comprises the following steps of; dividing and classifying types of enterprises and data units, extracting and collecting data of the enterprises after the enterprises are divided, analyzing the data after the data are collected, and adjusting the classified enterprises after the data are analyzed; step two, the analysis processing unit builds an enterprise selection model according to different enterprises through a large model technology, and builds a corresponding financial statement analysis scheme according to the classified enterprise types for analysis; And thirdly, carrying out risk early warning by the result processing unit according to the result after analysis and processing, and processing the risk after the risk early warning processing is completed.

Description

Automatic analysis system and method for enterprise financial report Technical Field The invention relates to the technical field of financial analysis, in particular to an automatic analysis system and method for enterprise financial reports. Background At present, an enterprise financial report is compiled based on accounting criteria, is a written document for reflecting financial conditions of a certain specific date, operation results and cash flow of an enterprise, is a core basis for decisions of interested parties such as investors, creditors, management layers and supervision institutions, and the like, and discloses an enterprise financial report automatic analysis system in China patent with a bulletin number of CN112950346B, which comprises a condition analysis setting module, a working state and analysis result output module, and an enterprise list acquisition module, a annual report acquisition module, an annual report data analysis module, a financial form acquisition module and an enterprise health condition assessment module which are respectively in business association with the condition analysis setting module, the working state and analysis result output module. Also discloses an automatic analysis method for the enterprise financial report. The invention realizes automatic acquisition of enterprise lists, enterprise annual reports, automatic extraction and arrangement of financial forms, automatic acquisition of key financial subject values of enterprises and automatic generation of comprehensive health condition assessment reports of enterprises, improves the efficiency of investors in analyzing enterprise annual reports, and reduces the difficulty of investors in analyzing enterprise annual reports. " The prior art only solves the problems that the existing evaluation process is long in time consumption, general investors do not have professional financial skills, a small amount of cash flow, profitability, growth capacity, operation capacity, repayment capacity and other health conditions of an enterprise to be invested can be obtained from investment tools, but in the existing financial analysis process, an analysis model cannot be formulated in advance according to the characteristic properties of the enterprise, so that the analysis processing efficiency is low, meanwhile, in the risk early warning process, the risk grade cannot be reasonably and accurately divided, even if the early warning generates risks, accurate and reasonable processing measures can not be carried out on the risk grade, and meanwhile, in the analysis process, repeated disc spot check processing cannot be carried out on analysis results and analysis results, so that the situation of human cheating can be caused. Disclosure of Invention The invention aims to provide an automatic analysis system and method for enterprise financial reports, which are used for solving the problems in the background technology. In order to achieve the aim, the invention provides the technical scheme that the enterprise financial statement automatic analysis system comprises an enterprise and data unit, an analysis processing unit and a result processing unit; The enterprise and the data unit divide and classify the types of the enterprise, extract and collect the data of the enterprise after the enterprise is divided, analyze the data after the collection is completed, and adjust the classified enterprise after the analysis; The analysis processing unit builds an enterprise selection model according to different enterprises through a large model technology, and builds a corresponding financial statement analysis scheme for analysis according to classified enterprise types; And the result processing unit performs risk early warning according to the analysis processed result after the analysis processing, and processes the risk after the risk early warning processing is completed. Preferably, the enterprise and data unit comprises an enterprise classification module, a data extraction module and a data analysis module, wherein the enterprise classification module classifies the enterprise in the following specific classification mode; (1) Classification according to legal forms, including individual business, partnership, finite liability company, stock finite company, national enterprises and foreign enterprises; (2) Classification by industry attribute including agriculture, forestry, animal husbandry, fishery, mining, manufacturing, electricity, heat, gas and water production, supply, construction, wholesale and retail industries, transportation, storage and postal industries, accommodation and catering industries, information transmission, software and information technology service industries, financial industries, residential industries, business service industries, and scientific research and technical service industries; (3) Classification by enterprise scale, including industry, retail, software and information technology