Application of three languages to pattern recognition
The purpose of this work is to develop new techniques in the theory of tree languages for applications in syntactic pattern recognition. The resulting procedures are presented as algorithms which are easy to implement in the form of computer programs.
The main areas covered are grammatical inference, minimization of tree automata, and the recognition of imperfect patterns. In Chapter II a new procedure is presented for the inference of tree grammars which is more general than existing methods. The minimization of tree automata is considered in Chapter V where a source of errors in the conventional approach is uncovered, and the method is modified to correct this problem. Finally, the recognition of imperfect patterns is studied in Chapter 6. The approach used to recognize imperfect patterns is based on an extension of tree languages to include trees specified by error operators. For this purpose a theory of regular-like expressions, representing tree languages, is developed.
Thesis78b.B379.pdf
1.36 MB
Adobe PDF
e52f11b58b6ba8a59fe87fe736417aec