Using Data Mining Algorithms to Predict Recommendations on Products

Abstract

Predict recommendations on products is important especially in understanding future customers’ ideas.Therefore, Prediction systems must beaccurate, effective and timely. This study predictsrecommendations on products by using data mining algorithms. Through using the WEKA data mining software tool, the C4.5 and Naïve Bayesalgorithms will be applied to the dress sales dataset. In an attempt to obtain better performance of the exported classifier model, Assigning a Unique Category (AUC) approach will be applied to handle missing data. The results show there is a rise in the accuracy of the classifier model after applying AUC missing data approach to missing values in the dataset.