Approach of information fusion and classification by SVM and DS evidence theory
Lei Le · Computer Engineering and Applications Journal · 2013
Based on the difficulty of obtaining the Basic Probability Assignment(BPA)of DS evidence theory in the practical application, an improved method of information fusion combing SVM and DS evidence theory is proposed. It uses the specific classification situation based on SVM and classifiers' reliabilities from confusion matrix to construct the basic probability assignment, which achieves the combination of SVM and the evidence theory in the information fusion. The method also presents a multi-sensor information fusion model. In the process of decision and fusion, it takes the sensors' local reliabilities into consideration and regards them as weights to integrate into BPA. The time complexity is also analyzed. The simulation results based on UCI data set and synthetic data set show that the fusion error rate can be decreased through the method proposed in this paper and the fusion reliabilities are increased.