Research And Implementation Of Decision Rule Classifier Based On SVM

As a kind of machine learning algorithm which is based on the statistics learning theoretics, the support vector machine (SVM) has excellent classification performance. In this paper, we introduce two boolean kernels. Using the kernel functions which we introduced, we built a decision rule classifier named as DRC-MBKF based on the linear mix of two boolean kernels, which is an understandable classification model by human experts. Experiment results show that, for both text classification and non-text classification, DRC-MBKF has more excellent classification performance than the DRC-BK. We also proposed a simplifyied algorithm by adopting constant iterative step length to optimize the parameter, which implemented selecting automatically. At last, a short-term load forecasting system was proposed based on DRC-MBKF, which has many merit such as good generalization, difficult to overfitting.