Authors : Raed Alazaidah, Farzana Kabir Ahmad and Mohamad Farhan Mohamad Mohsen
Abstract: In this study, a comparative analysis between the three main approaches that are being used to solve the problem of Multi Label Classification (MLC) have been conducted. The goal of doing this comparative analysis is to evaluate the performance of the three main approaches and decide which approach to use with respect to the characteristics of the multi label data set. Different from other comparative analysis that have been conducted in the field of MLC which focused on doing comparative analysis between different methods or algorithms, this study is focusing on the performance of the three main approaches that these methods and algorithms are belonging to. These three main approaches are: first, second and high order approaches. Results show that the high order approach is more suitable for solving the problem of MLC and has a better predictive performance than the other two approaches.
Raed Alazaidah, Farzana Kabir Ahmad and Mohamad Farhan Mohamad Mohsen, 2017. A Comparative Analysis Between the three Main Approaches that are being used to Solve the Problem of Multi Label Classification. International Journal of Soft Computing, 12: 218-223.