An improved multi-instance multi-label learning algorithm based on representative instances selection and label correlations. (2018)

• Record Type: Journal Article
• Authors: Liu, Chanjuan; Chen, Tongtong; Zou, Hailin; Ding, Xinmiao; Wang, Yuling
• Is Part Of: International journal of grid and utility computing. Volume 9:Number 3(2018)
• Page Start: 268
• Page End: 277
• Publication Date: 2018
• Subjects: multi-instance multi-label learning -- representative instances selection -- joint l2, 1-norms constraint -- label correlations; Electronic data processing -- Distributed processing -- Periodicals; Electronic commerce -- Management -- Computer programs -- Periodicals; 004.605
• Languages: English
• ISSNs: 1741-847X

Abstract

Multi-Instance Multi-Label Learning (MIML) has been successfully used in image and text classification problems. It is noteworthy that few of the previous studies consider the pattern-label relations. Inevitably, there are some useless instances in a bag which will reduce the accuracy of the annotation. In this paper we focus on this problem. Firstly, an instance selection method via joint l2, 1 -norms constraint is employed to eliminate the useless instances and select the representative instances by modelling the instance correlation. Then, bags are mapped to these representative instances. Finally, the classifier is trained by an optimisation algorithm based on label correlations. Experimental results on image data set, text data sets and bird song audio data set show that the proposed algorithm significantly improves the performance of MIML classifier compared with the state-of-the-art methods.