Litcius/Paper detail

TASTE

Ardavan Afshar, Ioakeim Perros, Haesun Park, Christopher R. deFilippi, Xiaowei Yan, Walter F. Stewart, Joyce C. Ho, Jimeng Sun

202036 citationsDOIOpen Access PDF

Abstract

nsor factorization (TASTE) that jointly models both static and temporal information to extract phenotypes. TASTE combines the PARAFAC2 model with non-negative matrix factorization to model a temporal and a static tensor. To fit the proposed model, we transform the original problem into simpler ones which are optimally solved in an alternating fashion. For each of the sub-problems, our proposed mathematical re-formulations lead to efficient sub-problem solvers. Comprehensive experiments on large EHR data from a heart failure (HF) study confirmed that TASTE is up to 14× faster than several baselines and the resulting phenotypes were confirmed to be clinically meaningful by a cardiologist. Using 60 phenotypes extracted by TASTE, a simple logistic regression can achieve the same level of area under the curve (AUC) for HF prediction compared to a deep learning model using recurrent neural networks (RNN) with 345 features.

Topics & Concepts

Computer scienceArtificial intelligenceNon-negative matrix factorizationMatrix decompositionRepresentation (politics)Machine learningLogistic regressionFactorizationDeep learningTensor (intrinsic definition)Pattern recognition (psychology)AlgorithmMathematicsEigenvalues and eigenvectorsPure mathematicsPoliticsPolitical scienceQuantum mechanicsLawPhysicsTensor decomposition and applicationsMachine Learning in Healthcare
TASTE | Litcius