Research On The Intelligent Evaluation Of Listed Companies' Accounting Information Quality

With the development of China's market economy,there are more and more listed companies,and the status of the securities market in the country has also increased.It also plays an increasingly important role.It is a barometer of the state of the economy.However,as the capital market continues to expand,many companies obtain high profits and money by disclosing false accounting information and fake transactions.At present,the phenomenon of financial fraud in the world is frequent,and the problem of accounting information distortion has become a topic of common concern all over the world.However,accounting information is used to record the accounting process and the results of the carrier,it can not only reflect the financial status of enterprises,but also reflect the business results and the flow of funds.In addition,it is an important basis that can be used to reflect the financial status of an enterprise,evaluate its performance,and decide whether to make a reproduction or investment decision.Now accounting information of listed companies will be publicly disclosed,so the accounting information of listed companies has become a public product.The level of accounting information of listed companies determines the formulation of national macroeconomic policy and the amount of citizens' wealth.On the one hand,the higher quality of accounting information helps accounting information demanders discriminate market risk,safeguard their own interests and make correct decisions.On the other hand,the higher quality of accounting information helps to effectively allocate resources and capital markets.Therefore,the quality of accounting information has become a hot issue for many scholars.On the evaluation of the quality of accounting information,many scholars are evaluated according to the eight characteristics of the quality of accounting information.In addition,the main research focuses on the influencing factors of accounting information quality.This paper mainly uses method of artificial neural network to quantify the influencing factors of accounting information quality of listed companies,and also studies the relationship among them,and assists with the methods of AHP and regression analysis.The research finds that the traditional methods of evaluating the quality of accounting information mainly include Shenzhen Stock Exchange disclosure quality,the quality of audit issued by the auditing firm,and the quality of accounting information based on the earnings quality.These three methods have their own advantages and disadvantages.The quality of the disclosure of the Shenzhen Stock Exchange is relatively strong.It comprehensively considers the accounting information voluntarily and forcibly disclosed by the listed companies.The quality of the disclosure is released by the authoritative agencies.The quality of disclosure is issued by an authoritative organization,which has a strong reliability and authority,and a strong degree of credibility.It is widely used.However,it is mainly based on the disclosure of listed companies to evaluate,and can't fully reflect the inherent quality of financial reports.The quality of the audit is the audit opinion issued by the accounting firm that the firm's external accounting firm issued based on the financial status and operating results of the company to the true reliability of the company's accounting information.However,even if auditors issued audit opinions based on relevant criteria,they are still affected by human factors and their objectivity needs to be improved.The correlation between earnings quality and accounting information quality is relatively strong,the model of the earnings quality is relatively mature,more applicability and evaluation index will obtain more easily,so many scholars will be based on earnings quality to evaluate the quality of accounting information.However,earnings quality is based on the established model to use the data disclosed by listed companies to infer the quality of accounting information.The result depends largely on the model and data obtained,so it can't well reflect the inherent quality of accounting information.Therefore,in this paper,AHP is used to evaluate the quality of accounting information by comprehensively considering the evaluation results,audit opinions and earnings quality of Shenzhen Stock Exchange.This result is more objective and accurate.From the characteristics of the board of directors,ownership structure,financial status and external factors in four aspects of the quality of accounting information quantitative analysis,these four indicators and then set up two or three small secondary indicators a total of 11 indicators,the use of good The BP neural network model directly scores the quality of accounting information and makes full use of the advantages of large-scale parallel,distributed processing,self-organization and self-learning of artificial neural network to make the final evaluation more accurate.Finally,BP neural network is used to get the result.The relative error between the two is controlled below 4% and the average error is 1.8%.It shows that the quality of accounting information can be objectively and accurately evaluated from the aspects of accounting information quality Evaluation results.It provides a new idea for the accounting information evaluation of listed companies.In this paper,the whole process of using the software Matlab2016,not only using the software to calculate the maximum eigenvalues and eigenvectors,but also with the software programming to build a three-layer BP neural network,to master the method for this article has played an important role.