Corpus-based Study Of Low-frequency Words In Contemporary English

This corpus-based research studies the distribution and characteristics of low-frequency words in naturally occurring texts quantitatively and qualitatively with the assistance of concordancing program.Statistical and computational results have shown that most of low-frequency words occurring in natural texts are open class words,comprising nouns,verbs,adjectives and adverbs;foreign origin words,particularly of Latin,Greek or French origin;wide ranges of technical terms;monosemic words,especially hyponyms denoting people,place and objects. Besides,from the viewpoint of quantity,there is regularity in the distribution of words of low frequencies. Considering words of frequency 1 to 10 as a whole:the number of low-frequency words (frequency=l) is the largest of all,the lower the frequency,the bigger the number of words with that frequency.The findings are of particular importance in wordlist compiling,which used to be compiled on the basis of the distribution of high-frequency words. Similarly,this research can shed light on vocabulary teaching and learning,allowing us to help students with the usage of these low-frequency words.