Classification of ATP on-board equipment test cases based on TF-IDF weighted Naive Bayesian algorithm

WANG Xinyi , CHENG Jianfeng , LIU Yujun Abstract: Aiming at the problems of heavy workload, low efficiency and low accuracy in the classification of test cases of on-board equipment of ATP (Automatic Train Protection) system, this paper proposed a scheme that combined TF-IDF (Term Frequency Inverse Document Frequency) with Naive Bayesian algorithm to classify test cases. The paper used TF-IDF algorithm to filter feature words and weights, and weighted Naive Bayesian algorithm, which was verified based on the test cases of existing ATP on-board equipment in the laboratory. The experiment results show that the method of feature word extraction and test case classification has high accuracy.