COMPRESSING TEXTS WITH NEURAL NETS¹

Jürgen Schmidhuber
IDSIA, Corso Elvezia 36
CH-6900-Lugano, Switzerland
http://www.idsia.ch/~ juergen
- Stefan Heil
Fakultät für Informatik
Technische Universität München
80290 München, Germany

Abstract:

Neural networks are promising tools for data compression without loss of information. We combine predictive neural nets and statistical coding techniques to compress text files. We apply our methods to certain short newspaper articles and obtain compression ratios exceeding those of widely used Lempel-Ziv algorithms (which are the basis of the UNIX functions ``compress'' and ``gzip''). Our methods' main disadvantage is that they are about three orders of magnitude slower than standard methods.