An outlier detection algorithm based on the degree of sharpness and its applications on traffic big data preprocessing

摘要: Outlier detection is one important research area of data mining, which plays key roles in preprocessing, equipment fault diagnosis, credit fraud detection, traffic incident etc. This paper devoted to a new outlier algorithm based on the degree sharpness. The proposed takes way solve problem, employs measure image processing, sharpness, detect outliers. Compared classical methods with statistical learning, has no iterative processes. It generates smooth curve describe overall distribution firstly, and then computes sharpness for each point. Finally, outliers are recognized as they have larger values Also, some practical applications big presented prove effectiveness algorithm.