Multi-view clustering via simultaneous weighting on views and features.pdfVIP

Applied Soft Computing 47 (2016) 304–315 Contents lists available at ScienceDirect Applied Soft Computing journal homepage: /locate/asoc Multi-view clustering via simultaneous weighting on views and features Bo Jiang a,?, Feiyue Qiu a, Liping Wang b a College of Education Science and Technology, Zhejiang University of Technology, Hangzhou 310023, China b College of Business and Administration, Zhejiang University of Technology, Hangzhou 310023, China article info Article history: Received 22 July 2015 Received in revised form 19 May 2016 Accepted 11 June 2016 Available online 17 June 2016 Keywords: Multi-view clustering Feature weighting View weighting abstract In big data era, more and more data are collected from multiple views, each of which re?ect distinct perspectives of the data. Many multi-view data are accompanied by incompatible views and high dimension, both of which bring challenges for multi-view clustering. This paper proposes a strategy of simultaneous weighting on view and feature to discriminate their importance. Each feature of multi-view data is given bi-level weights to express its importance in feature level and view level, respectively. Furthermore, we implements the proposed weighting method in the classical k-means algorithm to conduct multi-view clustering task. An ef?cient gradient-based optimization algorithm is embedded into k-means algorithm to compute the bi-level weights automatically. Also, the convergence of the proposed weight updating method is proved by theoretical analysis. In experimental evaluation, synthetic datasets with varied noise and missing-value are created to investigate the robustness of the proposed approach. Then, the proposed approach is also compared with ?ve state-of-the-art algorithms on three real-world datasets. The experiments show that the proposed method compares very favourably against the other methods. ? 2016 Elsevier B.V. All rights reserved. 1. Introduction Multi-view clustering is concerned with the pr