Introduction

Abstract

The extensive application and the substantial further development of pattern recognition methods on the basis of Markov models took place in the field of automatic speech recognition. There the combination of hidden Markov models for the acoustic analysis and Markov chain models for the restriction of potential word sequences is the predominant paradigm today. In contrast, their use in different application areas such as, for example, character or handwriting recognition or the analysis of biological sequences, becomes accessible from the respective specialized technical literature only. This is surprisingly true also for the presentation of Markov chain models which are usually referred to as statistical language models. The situation is the same for questions which arise in combination with the practical application of Markov model technology. Therefore, this book pursues two goals. First, Markov models will be presented with respect to their nowadays extremely wide application context. Secondly, the treatment will not be concentrating on the theoretical core of the modeling only, but will include all technological aspects that are relevant from today’s view.