An attribute reduction SVM-based tax assessment model

Tax assessment has always been regarded as a tool to check whether taxpayers submit the right amount of money. This paper propose a new tax assessment model mainly based on an attribute reduction SVM in the light of its good performance in classifying high-dimensional nonlinear data. In the paper, we apply our model to a real-world example to see its practical performance. The results show that our model performs well both in data classification accuracy and predictive accuracy.