A fuzzy recommender system for forecasting customer segmentation by multi-variable fuzzy rule interpolation

This work is presenting a method to assist customers with their purchasing enquiries using fuzzy forecasting method by segmenting them to specific groups. After customer segmentation, we could forecast customer requirements. Customer Segmentation is based on customer´s requirements and characteristics which cannot be categorized in an absolute or specific way as there is fuzziness in customers´ personal and their requirements and product specifications. Fuzzy set theories are employed due to the presence of vagueness and imprecision of information in our problem. For this purpose, Gaussian membership function is chosen because it demonstrates a smoother transition in its intervals, and the achieved results were closer to the actual ones. The contribution of this paper is that it applies one of the most recent existing achievements in multi-variable fuzzy forecasting for anticipating customer requirements. Our new methodology has two parts; a fuzzy clustering based on fuzzy C-means for customer segmentation and a fuzzy forecasting method by multi-variable rule interpolation technique. At the end, we could suggest customers the products which satisfy them.