Outlier detection method based on SVM and its application in copper-matte converting

Outlier detection can be treated as a part of the data preprocess or as the object of data mining. There is still no effective detection method for the high-dimensional nonlinear outlier samples. This paper presents an outlier detection method based on support vector machine (SVM). A SVM model built by the clean sample set without outlier is used to predict the samples, when the error between the prediction-value and actual value exceeds the threshold, the sample is taken as an outlier, otherwise a normal one. The present outlier detection method has been applied to analyze the practical copper-matte converting production data. The results show that this method can efficiently and correctly detect the high dimensional nonlinear outlier sample and has considerable practical value.