Fuzzy Soft Set Euclidian Based for Data Classification

Nurfitri Ani, Ririn Setiyowati, Iwan Tri Riyadi Yanto

This study discusses about data classification using fuzzy soft set method which be used to predict target classes accurately. The purpose of this study is to form data classification algorithm using fuzzy soft set method. In this method, each parameter is mapped to power set from subset of fuzzy set using fuzzy approximation function. There are two algorithms in data classification, such as training and classification step. In the classification step, a Euclidean distance is used to determine the similarity between two sets of fuzzy soft sets. Data classification with the fuzzy soft set method is implemented to UCI Machine Learning data. Eventually, computational experiment show that the result are 92.92%, 92.92% and 89.37% for accuracy, Fmicro and Fmacro, respectively. It means that the fuzzy soft set method is appropriate to be uses for classify data.

Volume 12 | 02-Special Issue

Pages: 679-685

DOI: 10.5373/JARDCS/V12SP2/SP20201121