Computer linguistics and language technology. An introduction (Q2756799)

The book is structured as a compendium of introductions into various topics of computation linguistics (CL). The topics are grouped into six chapters (1. What is CL, 2. Formal Basics, 3. Methods, Resources, 4. Applications, and 5. Evaluation of CL Systems). Each chapter has its editor. As a rule each section is written by its own author(s) (the book contains contributions of 41 authors) and contains hints for further reading. Chapter 2 (pages 25-132) contains short introduction into the theory of sets and logics into the theory of automata and formal languages, and into probability theory and Markov models. The next chapter (pages 135-360) contains sections dealing with phonology, morphology, syntax and parsing, semantics, discourse and pragmatics, and text generation. The chapter Resources discusses WWW, text corpora, treebanks, lexical-semantic word networks, and lexis for multimodal systems. The chapter Applications (409-521) discusses e.g. spelling checking programs, various full text systems, speech generation and speech recognition, dialog systems, computer aided language translation. NEWLINENEWLINENEWLINEIt is obvious that the introductions contained in the book are very short, sometimes like extended annotations. The best way how to use the book is to use it as a source (dictionary) of the basic definitions and results and the source of references. Some sections can be also used as shortened introductions. The structure of the book implies the importance of references and literature links towards resources allowing to achieve a deeper knowledge. The links in the book are classical references only. There are no hints how to access the latest results via www (e.g. keywords in German and English and/or sources on www) and no list of relevant journals and leading conferences. The domain of CL is very broad. Some topics must be therefore omitted. What ones is a matter of taste. Some omissions are, however, quite important. The book does not discuss the methods based on mathematical statistics and the dependency grammars important not only for the free word order languages.