Abnormal Data Monitoring and Analysis Based on Data Mining and Neural Network

In order to solve the problems of low efficiency, high consumption human and time resources, degree intelligence in current financial abnormal data detection system computerized accounting, this paper proposes a monitoring analysis algorithm based on mining neural network. The uses method process original data, remove invalid information, retain valuable standardize problem large labor consumption. Then, network-related algorithms are used identify anomalies standardized so as realize intelligent early warning data. Compared with audit traditional accounting computerization, has advantages energy consumption, intelligence. test results show that classification accuracy proposed for can reach more than 90%. It is proved effective improves efficiency at same time. error rate classifier designed 22.5%, 77.5%. Both estimated actual values represent number times, there no physical unit. experiment shows main reason delay inspection abnormalities. Through example analysis, it be concluded deep learning good effectiveness certain practical value.