Neural Network Technology In Use In The Evaluation Of Internal Control

At the end of the twentieth century, some financial frauds of famous worldwide companies have erupted around the world. For these, some countries have issued relevant laws and regulations. The COSO report and the SOX act have provided the guidances for establishing and perfecting the internal control for the enterprises and have also caused the studies of internal control and its evaluation. The ancients said " A good doctor cures a disease that not happens, an ordinary doctor heals a disease that is about to happen, a poor doctor cures a disease that has already taken place". Enterprise internal control evaluation, such as "a good doctor cures a disease that not happens ". The design aims to make people know where exist defects and which aspect may lead to consequences and then take actions to change this situation by themselves rather than waiting for the loss caused by hazards developing. During2013the Green Land fraud and Everbright Securities "mistake" event have caused serious consequences because of the lack or failure of internal control, and increasingly emphasized the importance of constructing internal control and its evaluation system.This paper mainly researches neural network technology uses in the evaluation of internal control. Internal control evaluation as the important means to perfect internal control system and ensure the effective implementation is a necessary part of internal control objectives. Meantime the research and application of neural network is under way. It has a huge attraction for artificial intelligence research with a whole new way of expressing and processing information. It gets good results in pattern recognition, combinatorial optimization, decision and making judgments, evaluation and prediction that traditional computer is difficult to achieve. So neural network technology which is applied to the internal control evaluation has its unique advantages in the rationality and applicability about building internal control evaluation models.This article first introduces the studies of enterprise internal control evaluation at home and abroad and the research status of neural network technology. Then, it has carried on the simple introduction about internal control and its evaluation, so does neural network theories. After the analysis of internal control evaluation theory framework, Including the establishment of evaluation system of internal control and its principles, scoring criteria and evaluation results were analyzed. Finally, by adopting the combination of qualitative and quantitative methods we build the evaluation index system of internal control, we evaluate the internal control through fuzzy comprehensive evaluation method and BP neural network model quantitativing the results of qualitative evaluation.In this paper, the research on the construction and evaluation of enterprise internal control system is of great realistic significance. It both improves the enterprise internal control evaluation theory model and it is conductive for listed companies to implement the basic norms from the application perspectives. And it is beneficial for listed company to correctly evaluate the quality of internal control operation, understanding and analysis of problems during the execution of internal control, and take effective measures to solve these problems. In addition, the internal control evaluation system can also act as a "thermometer", it can timely evaluate and make a diagnosis of the internal control status and improve corporate performance, the prosperity of our country capital market by protecting the interests of the investors so as to promote the healthy and orderly development of our country’s economy. In spite of this, this article is just using neural network technology in the study of internal control evaluation of a starting point, the research needs to be further expanded in the future.