CLASSIFICATION MODELS BASED FORWARD SELECTION FOR BUSINESS PERFORMANCE PREDICTION

NOERSASONGKO, EDI and -, PURWANTO and SHIDIK, GURUH FAJAR (2014) CLASSIFICATION MODELS BASED FORWARD SELECTION FOR BUSINESS PERFORMANCE PREDICTION. Document Repository.

Abstract

This paper proposes a classification model to improve the accuracy of prediction for business performance. The proposed model uses a combination of forward selection method to select the optimum attributes and classification models. Business performance data set is used to evaluate the accuracy of the proposed model. From results of experiments show that the combination of forward selection and Naïve Bayes model can improve the prediction accuracy of business performance compared to the other classification models, namely Logistic Regression, k-NN, Naïve Bayes, C4.5 and Support Vector Machine models significantly. The proposed model also yields better result compared to the other attribute selection using backward elimination method. Keywords: Forward Selection, Naïve Bayes, Entrepreneur, Business performance, Classification Models

Actions (login required)