BFA and BMF: What is the Difference

0382263 - ÚI 2013 RIV US eng C - Konferenční příspěvek (zahraniční konf.)
Frolov, A. A. - Abraham, A. - Polyakov, P.Y. - Húsek, Dušan - Řezanková, H.
BFA and BMF: What is the Difference.
Intelligent Systems Design and Applications (ISDA). Piscataway: IEEE Computer Society, 2012 - (Abraham, A.; Zomaya, A.; Ventura, S.; Yager, R.; Snášel, V.; Muda, A.; Samuel, P.), s. 890-896. ISBN 978-1-4673-5117-1. ISSN 2164-7143.
[ISDA 2012. International Conference on Intelligent Systems Design and Applications /12./. Kochi (IN), 27.11.2012-29.11.2012]
Grant CEP: GA ČR GAP202/10/0262
Grant ostatní: GA MŠk(CZ) ED1.1.00/02.0070
Program: ED
Institucionální podpora: RVO:67985807
Klíčová slova: dimension reduction * statistics * data mining * Boolean factor analysis * Boolean matrix factorization * information gain * likelihood-maximalization * bars problem
Kód oboru RIV: IN - Informatika

Studied are differences of two approaches to binary data dimension reduction. The first one is Boolean Matrix Factorization and the second one is Expectation Maximization Boolean Factor Analysis. The two BMF methods are used for comparison. First is M8 method from the BMDP statistical software package. The second is the BMF method, as suggested by Belohlavek and Vychodil [1]. These two are compared to Expectation Maximization Boolean Factor Analysis extended with binarization step developed here. Generated (Bars problem) and mushroom dataset are used for experiments. In particular, under scrutiny was the reconstruction ability of the computed factors and the information gain as the measure of dimension reduction. In addition, presented are some general remarks on all the methods being compared.
Trvalý link: http://hdl.handle.net/11104/0212537