Corpus Analysis of Specialised Vocabulary and Polysemy in the Logistics Academic Written English

This research focuses on a specific discipline, Logistics, to explore specialised vocabulary and polysemy in the Logistics academic written context through corpus. Results of this research were generated from a self-built specialised corpus of 5.35 million tokens on Logistics academic written English, covering 720 texts sampled from nine top academic journals in that field.
To select potential words for polysemy analysis, a Logistics specialised academic written English vocabulary list (the LoSAWEL) was developed through keyness, range, dispersion, focusing on lexical words. This list contains 311 individual word types, each classified at different levels with Nation’s (2012) BNC/COCA 25 base word lists as references to build connection with their general use frequency.
Word sense identification (collocation analysis, concordance line reading, and general use and discipline-specific dictionary checking) was conducted in the LAWEC (a self-built specialised corpus on Logistics academic written English), the BAWE (a general academic corpus) and the BNC1994 (a general use corpus) respectively. Word sense comparisons were conducted between the Logistics specialised context and general academic context, and between the Logistics specialised context and general use context. Results show that about half of senses had difference in senses in each round of word sense comparison respectively. Adverb and preposition senses were used consistently across corpora. Noun, verb and adjective senses show differences across registers. Shared senses between the Logistics specialised context and general academic context suggest the potential existence of a group of core academic vocabulary, and the difference of senses show disciplinary variation of word sense and word use. Also, register plays a role on differences of word senses.
Results also showed that more than 1/3 individual words in the LoSAWEL are polysemous words. These polysemous words have relations with their levels of general use frequency. The higher general use frequency level a word has, the more likely it can be a polysemous word. These words are also associated with part of speech, with adverbs and prepositions less likely to be polysemous words but nouns, verbs, and adjectives are. This research will contribute to corpus analysis on polysemy in specific context.