[2P3] [General Session] 2. Machine Learning

[2P3-05] Multi-label Logistic Regression Using Relative Density Ratio

〇Masaaki Okabe¹, Jun Tsuchida², Hiroshi Yadohisa³ (1. Graduate School of Cluture and Infomation Science, Doshisha University, 2. Faculty of Engineering, Tokyo University of Science, 3. Faculty of Culture and Information Science, Doshisha University)

Keywords:imbalanced data, Binary relebance, F-measure

Multi-label classification is a supervised learning problem where multiple labels may be assigned to each instance. The main baseline for multi-label classification is binary relevance method, which is estimate the binary classification model for each label. In binary classification, there are cases where poor results are data when the class is imbalance.
In this paper, we propose a multi-label classification model used relative density ratio. In this model, we used relative F-measure by relative density ratio for weight of error function to solve the class imbalance problem.