Incorporating Multiple Cluster Centers for Multi-Label Learning

Multi-label learning deals with the problem that each instance is associated with multiple labels simultaneously. Due to its ability to cope with the real-world objects with multiple semantic meanings, multi-label learning has been successfully applied in various application domains [1], such as tag recommendation [2, 3], bioinformatics [4, 5, 6], information retrieval [7, 8], rule mining [9, 10], web mining [11, 12], and so on. Formally speaking, suppose the given multi-label data set is denoted by D {x i, y i } n i 1 where x i R d is a feature vector with d dimensions (features) and y i { 1, 1} q is the corresponding label vector with the size of label space being q. Here, y ij 1 indicates that the i-th instance x i has the j-th label (or equivalently, the j-th label is a relevant label of x i), otherwise the j-th label is an irrelevant label of x i . Let X R d be the d-dimensional feature space, and Y { 1, 1} q be the q-dimensional label space, multi-label learning aims to induce a mapping function f: X Y, which is able to correctly predict the label vector of unseen instances. To solve the multi-label learning problem, the most straightforward solution is Binary Relevance (BR) [13, 14], which aims to decompose the original learning problem into a set of independent binary classification problems. However, this solution generally achieves mediocre performance, as label correlations are regrettably ignored. To ease this problem, a large number of multi-label learning approaches take into account label correlations explicitly or implicitly to improve the learning performance.