MMMF: Multimodal Multitask Matrix Factorization for Classification and Feature Selection
Open Access
- 10 November 2022
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 120155-120167
Abstract
Integration of multiple biological datasets is crucial to understand comprehensive biological mechanisms with the aid of a rapid development of biomedical technology. However, the predictive modeling for such an integrated dataset faces two major challenges, namely, heterogeneity and imbalance in the acquired data. Thus, in this study, we present a method for the integration of multiple biological datasets called multimodal multitask matrix factorization (MMMF) to address these issues. The MMMF uses matrix factorization (MF) to integrate data from multiple heterogeneous biological datasets, and oversampling is applied to resolve the imbalanced data during the training step. Moreover, gradient surgery is used for multitask (MF and classification) learning to increase the quantity of classification information by projecting the gradients of the MF that conflict with the classification gradient onto the normal plane of a classification gradient. We demonstrate that MMMF outperforms other state-of-the-art biomedical classification models in binary and multi-class classification problems using five biological datasets. We also show that MMMF can be used as a feature selection approach for finding biomarkers that help in classification. The source code of the MMMF is available at https://github.com/DMCB-GIST/MMMF.Keywords
Funding Information
- Bio & Medical Technology Development Program of NRF funded by the Korean Government [Ministry of Science and ICT (MSIT)] (NRF-2018M3C7A1054935)
- Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government (MSIT) (2019-0-00567)
- Artificial Intelligence Graduate School Program, Gwangju Institute of Science and Technology (2019-0-01842)
This publication has 41 references indexed in Scilit:
- Variable selection for generalized canonical correlation analysisBiostatistics, 2014
- Similarity network fusion for aggregating data types on a genomic scaleNature Methods, 2014
- Relationship between imaging biomarkers, age, progression and symptom severity in Alzheimer's diseaseNeuroImage: Clinical, 2013
- CLASSIFICATION OF IMBALANCED DATA: A REVIEWInternational Journal of Pattern Recognition and Artificial Intelligence, 2009
- SVD based initialization: A head start for nonnegative matrix factorizationPattern Recognition, 2008
- The Alzheimer's disease neuroimaging initiative (ADNI): MRI methodsJournal of Magnetic Resonance Imaging, 2008
- A Phosphoproteomic Analysis of the ErbB2 Receptor Tyrosine Kinase Signaling PathwaysBiochemistry, 2006
- Association of DSC1, a gene modulated by adrenergic stimulation, with Alzheimer's diseaseNeuroscience Letters, 2006
- Adjusting batch effects in microarray expression data using empirical Bayes methodsBiostatistics, 2006
- Generalized Linear ModelsJournal of the Royal Statistical Society. Series A (General), 1972