Boolean Factor Analysis for Data Preprocessing in Machine Learning

We present two input data preprocessing methods for machine learning (ML). The first one consists in extending the set of attributes describing objects in input data table by new attributes and the second one consists in replacing the attributes by new attributes. The methods utilize formal concept analysis (FCA) and boolean factor analysis, recently described by FCA, in that the new attributes are defined by so-called factor concepts computed from input data table. The methods are demonstrated on decision tree induction. The experimental evaluation and comparison of performance of decision trees induced from original and preprocessed input data is performed with standard decision tree induction algorithms ID3 and C4.5 on several benchmark datasets.