Using Blind Search and Formal Concepts for Binary Factor Analysis.pdf

Using Blind Search and Formal Concepts for Binary Factor Analysis Aleˇs Keprt Dept. of Computer Science, FEI, VSB Technical University Ostrava, Czech Republic ales.keprt@vsb.cz Abstract. Binary Factor Analysis (BFA, also known as Boolean Factor Analysis) may help with understanding collections of binary data. Since we can take collections of text documents as binary data too, the BFA can be used to analyse such collections. Unfortunately, exact solving of BFA is not easy. This article shows two BFA methods based on exact computing, boolean algebra and the theory of formal concepts. Keywords: Binary factor analysis, boolean algebra, formal concepts 1 Binary factor analysis 1.1 Problem definition To describe the problem of Binary Factor Analysis (BFA) we can paraphrase BMDP’s documentation (Bio-Medical Data Processing, see [1]). BFA is a factor analysis of dichotomous (binary) data. This kind of analysis differs from the classical factor analysis (see [16]) of binary valued data, even though the goal and the model are symbolically similar. In other words, both classical and binary analysis use symbolically the same notation, but their senses are different. The goal of BFA is to express p variables X = (x , x , . . . , x ) by m factors 1 2 p (F = f , f , . . . , f ), where m p (m is considerably smaller than p ). The 1 2 m model can be written as X = F A where is matrix multiplication. For n cases, data matrix X , factor scores F , and factor loadings A can be written as   