In this paper we compare the performance of several dimension reduction techniques which are used as a tool for feature extraction. The tested methods include singular value decomposition, semi-discrete decomposition, non-negative matrix factorization, novel neural network based algorithm for Boolean factor analysis and two cluster analysis methods as well. So called bars problem is used as the benchmark. Set of artificial signals generated as a Boolean sum of given number of bars is analyzed by these methods. Resulting images show that Boolean factor analysis is upmost suitable method for this kind of data. © 2008 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Snášel, V., Moravec, P., Húsek, D., Frolov, A., Řezanková, H., & Polyakov, P. (2008). Pattern discovery for high-dimensional: Binary datasets. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 4984 LNCS, pp. 861–872). https://doi.org/10.1007/978-3-540-69158-7_89
Mendeley helps you to discover research relevant for your work.