An Aggregated Fuzzy Naive Bayes Data Classifier

In this study, an Aggregated Fuzzy Naive Bayes Classifier is proposed for decision-making problems where both linguistic and numerical information are available. In the solution process of such problems, all attributes are considered as fuzzy numbers and a procedure based on 2-tuple fuzzy linguistic representation model is generated for combining them. This procedure and subsequent Fuzzy Naive Bayes classification are performed based on arithmetic operations defined by Chen’s function principle. The proposed method was demonstrated on 2 well-known examples from the literature in which both numerical and linguistic attributes were considered. The results show that the proposed Aggregated Fuzzy Naïve Bayes Classifier is notably efficient in decision-making where the attributes are in more realistic forms.