• RMIT Australia
  • RMIT Global
  • RMIT Australia
  • RMIT Europe
  • RMIT Vietnam

• Students
• Alumni
• Staff

Research Repository

Back to Library

• Home
• Browse
• Search
• FAQs
• About
• Contact Us

A Bayesian classifier for learning from tensorial data Liu, W, Chan, J, Leckie, C, Chen, F and Ramamohanarao, K 2013, 'A Bayesian classifier for learning from tensorial data', in H. Blockeel, K. Kersting, S. Nijssen, F. Zelezný (ed.) Proceedings of European Conference on Machine Learning and Pinciples and Practices of Knowledge Discovery in Databases (ECML PKDD) 2013, Prague, Czech Republic, 23-27 September 2013, pp. 483-498. Document type: Conference Paper Collection: Conference Papers Title A Bayesian classifier for learning from tensorial data Author(s) Liu, W Chan, J Leckie, C Chen, F Ramamohanarao, K Year 2013 Conference name ECML PKDD: European Conference on Machine Learning and Pinciples and Practices of Knowledge Discovery in Databases Conference location Prague, Czech Republic Conference dates 23-27 September 2013 Proceedings title Proceedings of European Conference on Machine Learning and Pinciples and Practices of Knowledge Discovery in Databases (ECML PKDD) 2013 Editor(s) H. Blockeel, K. Kersting, S. Nijssen, F. Zelezný Publisher Springer Place of publication Germany Start page 483 End page 498 Total pages 16 Abstract Traditional machine learning methods characterize data observations by feature vectors, where an entry of a vector denotes a scalar feature value of a data instance. While this data representation facilitates the application of conventional machine learning algorithms, in many cases it is not the best way of extracting all useful information from the data observations. In this paper we relax the (often unstated) assumption of vectorizing features of data instances, and allow a more natural representation of the data in a tensor format. Tensors are multi-mode (aka multi-way) arrays, of whom vectors (i.e., one-mode tensors) and matrices (i.e., two-mode tensors) are special cases. We show that the tensor representation captures useful information that is difficult to provide in the conventional vector format. More importantly, to effectively utilize the rich information contained in tensors, we propose a novel semi-naive Bayesian tensor classification method (which we call Bat) that builds predictive models directly on data in tensor form (instead of on their vectorizations). We apply Bat to supervised learning problems, and perform comprehensive experiments on classifying text documents and graphs, which demonstrate (1) the advantage of the tensor representation over conventional feature-vectorization approaches, and (2) the superiority of the proposed Bat tensor classifier over other existing learners. Subjects Pattern Recognition and Data Mining DOI - identifier 10.1007/978-3-642-40991-2 Copyright notice © Springer ISBN 9783642409912 Related Links Link Description http://dx.doi.org/10.1007/978-3-642-40991-2 Link to Published Version   Versions Version Filter Type Thu, 30 Apr 2015, 09:45:09 EST Filtered Full Citation counts: Cited 0 times in Scopus Article Altmetric details: Access Statistics: 273 Abstract Views  -  Detailed Statistics Created: Thu, 30 Apr 2015, 08:42:00 EST by Catalyst Administrator