Computational Approach for Processing of Control Engineer Text: Applications for Corpus Lexicography

Abstract

This research project reflects an increased awareness of the need to improve the flow of information on control engineering technology and the current lack of bilingualized (English-Thai-English) dictionary in this field. The central aim of this project is to develop a corpus of control engineering text. It starts with a survey of users' needs in control engineering English from 3 areas, i.e. control systems, automation and instrumentation, extracted from textbooks and journals. The corpus comprises 2,141,293 words (tokens) of running text. The linguistic data of the corpus provides insights into the sublanguage of control engineering. The corpus was then analysed by concordance program named WordSmith Tools Package to discover the frequency list of individual words, and collocations. The corpus findings in terms of word frequencies, corpus evidence on word combinations and typical usage from the concordance are used to find out the technical terms and to develop 665 entries for the bilingualized learners' dictionary of control engineering