Performance of Classification Techniques along with Support Vector Machine

Abstract

Statistical learning is one of the most notable fields studied by the researchers to understand the data in the present scenario. Recent advances in the field of machine learning and artificial intelligence have been keen to develop more powerful automated techniques for predictive modeling, specifically in regression and classification models. These approaches fall under supervised statistical learning techniques, many conventional techniques are very complex to the data when it has larger volumes, i.e., if the data deviates from the model assumption, then the conventional procedure’s results does not have the trustworthy. This paper explores and compares the classical methods with the alternatives in the context of classification, like logistic regression and support vector machine. The efficiency of these procedures has been evaluated through various measures such as confusion matrix and misclassification rate under real environment.