Continuous Semi-Supervised Nonnegative Matrix Factorization
Computation and Language
2022-12-21 v1 Machine Learning
Abstract
Nonnegative matrix factorization can be used to automatically detect topics within a corpus in an unsupervised fashion. The technique amounts to an approximation of a nonnegative matrix as the product of two nonnegative matrices of lower rank. In this paper, we show this factorization can be combined with regression on a continuous response variable. In practice, the method performs better than regression done after topics are identified and retrains interpretability.
Cite
@article{arxiv.2212.09858,
title = {Continuous Semi-Supervised Nonnegative Matrix Factorization},
author = {Michael R. Lindstrom and Xiaofu Ding and Feng Liu and Anand Somayajula and Deanna Needell},
journal= {arXiv preprint arXiv:2212.09858},
year = {2022}
}