Authors : Mo`ath Al-luwaici, Ahmad Kadri Junoh and Farzana Kabir Ahmad
Abstract: In the last few decades, fuzzy logic has been extensively used in several domains such as economy, decision making, logic and classification. In specific, fuzzy logic which is a powerful mathematical representation has shown a superior performance with uncertainty real-life applications comparing with other learning approaches. Many researchers utilized the concept of fuzzy logic in solving the traditional single label classification problems of both types: binary classification and multi-class classification. Unfortunately, very few researches have utilized fuzzy logic in a more general type of classification that is called Multi-Label Classification (MLC). Hence, this study aims to examine the applicability of fuzzy logic to be used with MLC through evaluating several fuzzy-based classifiers on five different multi-label datasets. The results revealed that the utilizing fuzzy-based classifiers on solving the problem of MLC is promising comparing with a wide range of MLC algorithms that belong to several learning approaches and strategies.
Mo`ath Al-luwaici, Ahmad Kadri Junoh and Farzana Kabir Ahmad, 2019. Investigating the Applicability of Several Fuzzy-Based Classifiers on Multi-Label Classification. Journal of Engineering and Applied Sciences, 14: 7210-7217.