Research and Improvement of Internet Financial Anti-Fraud Rules Based on Information Gain and Support

Abstract

Using rules to develop internet financial anti-fraud is one of the important means in this field. How to improve the existing rules and give better perform to their effectiveness is a hotspot in the field of internet financial anti-fraud. Based on the FOIL information gain, Gini coefficient, and support degree, this paper researches how to improve the internet financial anti-fraud rules with the combination of rules learning and improving technology. With the combination of these indexes, this paper analyzes the experimental results with using the real transaction data of banks. The experimental results show that the improved rules have good performance in support, and have some improvement for the internet financial anti-fraud.