BFA and BMF: What is the difference

Author

Frolov, A.A. ; Abraham, Ajith ; Polyakov, P.Y. ; Husek, D. ; Rezankova, H.

Author_Institution

Inst. of Higher Nervous Activity, Moscow, Russia

fYear

2012

fDate

27-29 Nov. 2012

Firstpage

890

Lastpage

896

Abstract

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 (BVA2). 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.

Keywords

Boolean algebra; data mining; matrix decomposition; optimisation; software packages; statistical analysis; BFA; BMDP statistical software package; BMF methods; Boolean matrix factorization; M8 method; binarization step; binary data dimension reduction; dimension reduction; expectation maximization Boolean factor analysis; information gain; mushroom dataset; reconstruction ability; Algorithm design and analysis; Bars; Electronic mail; Intelligent systems; Loading; Matrix decomposition; Vectors; Boolean factor analysis; Boolean matrix factorization; Dimension reduction; bars problem; data mining; information gain; likelihood-maximization; statistics;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems Design and Applications (ISDA), 2012 12th International Conference on

Conference_Location

Kochi

ISSN

2164-7143

Print_ISBN

978-1-4673-5117-1

Type

conf

DOI

10.1109/ISDA.2012.6416656

Filename

6416656