Identifying Abnormal Corporate Financial Data Based on the Comparison of SVM and Logistic Algorithms

With the continuous development of Chinese market economy, Chinese enterprises are facing more and more complicated and fierce challenges if they want to get new opportunities for development in the market. In the market economy environment, the competition between enterprises is fierce. This paper aims to study the abnormal financial data identification based on SVM and Logistic algorithm. Through the research results of SVM, Logistic algorithm and financial identification index at home and abroad, the idea of constructing this model is clarified. In the construction of the enterprise financial abnormal data identification model, 90 enterprises as research objects to identify financial anomalies. The experimental results show that the recognition results are stable and the accuracy is high with SVM modeling, but the accuracy is decreased after the addition of non-financial data. Compared with SVM, Logistic has a lower accuracy in this problem.