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 h 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.