An Attribute Reduction Method Based on Rough Set and SVM and with Application in Oil-Gas Prediction

With greater generalization performance Support Vector Machine (SVM) is a new machine learning method. Rough Set Theory is a new powerful tool in dealing with vagueness and uncertainty information. By combining the advantages of two approaches, an original attribute reduction method is proposed in the paper. Moreover, it is applied into oil-gas prediction to solve the problems when support vector machine is directly employed. Experiments and results show the validity and feasibility of the algorithm suggested in the paper.