Turbo‐SMT: Parallel coupled sparse matrix‐Tensor factorizations and applications. (30th June 2016)

• Record Type: Journal Article
• Authors: Papalexakis, Evangelos E.; Mitchell, Tom M.; Sidiropoulos, Nicholas D.; Faloutsos, Christos; Talukdar, Partha Pratim; Murphy, Brian
• Is Part Of: Statistical analysis and data mining. Volume 9:Number 4(2016)
• Page Start: 269
• Page End: 290
• Publication Date: 2016-06-30
• Subjects: algorithm -- coupled matrix‐tensor factorization -- fMRI data -- neurosemantics -- parallel -- sparse -- speedup -- tensor; Data mining -- Statistical methods -- Periodicals; 006.312
• DOI: 10.1002/sam.11315
• Languages: English
• ISSNs: 1932-1864

Abstract

Abstract : How can we correlate the neural activity in the human brain as it responds to typed words, with properties of these terms (like 'edible', 'fits in hand')? In short, we want to find latent variables, that jointly explain both the brain activity, as well as the behavioral responses. This is one of many settings of the Coupled Matrix‐Tensor Factorization (CMTF) problem. Can we enhance any CMTF solver, so that it can operate on potentially very large datasets that may not fit in main memory? We introduce Turbo ‐SMT, a meta‐method capable of doing exactly that: it boosts the performance of any CMTF algorithm, produces sparse and interpretable solutions, and parallelizes any CMTF algorithm, producing sparse and interpretable solutions (up to 65 fold ). Additionally, we improve upon ALS, the work‐horse algorithm for CMTF, with respect to efficiency and robustness to missing values. We apply Turbo ‐SMT to Brain Q, a dataset consisting of a (nouns, brain voxels, human subjects) tensor and a (nouns, properties) matrix, with coupling along the nouns dimension. Turbo ‐SMT is able to find meaningful latent variables, as well as to predict brain activity with competitive accuracy. Finally, we demonstrate the generality of Turbo ‐SMT, by applying it on aFACEBOOK dataset (users, 'friends', wall‐postings); there, Turbo ‐SMT spots spammer‐like anomalies. © 2016 Wiley Periodicals, Inc. Statistical Analysis and Data Mining: The ASA Data Science Journal, 2016