Abstract
Boolean matrix decomposition is a method to obtain a compressed representation of a matrix with Boolean entries. We present a modular framework that unifies several Boolean matrix decomposition algorithms, and provide methods to evaluate their performance. The main advantages of the framework are its modular approach and hence the flexible combination of the steps of a Boolean matrix decomposition and the capability of handling missing values. The framework is licensed under the GPLv3 and can be downloaded freely at http://projects.informatik.uni-mainz.de/bmad. © 2014 Springer-Verlag.
Cite
CITATION STYLE
Tyukin, A., Kramer, S., & Wicker, J. (2014). BMaD - A Boolean matrix decomposition framework. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8726 LNAI, pp. 481–484). Springer Verlag. https://doi.org/10.1007/978-3-662-44845-8_40
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
