| LSA, Latent Semantic Analysis, is a fully automatic computer method for the construction of a knowledge representation in the form of a high-dimensional semantic space based on the analysis of a large corpus of written text. The computer reads a large amount of word tokens-consisting of thousands of documents and word types. LSA is an approach which can perform semantic matching that can be applied to a variety of areas in text comprehension research that typically use modeling with propositions. The method is proved to be automatic and fast, permitting quick measurements of the semantic similarity between pieces of textual information.In recent years, writing, as one of the most important communicative skills, is drawing more and more attention in English teaching in China. In this paper, an investigation is carried out, by using decomposition in LSA and linguistic statistical methods, to analyze the textual coherence. Based on LSA 's measuring textual coherence and Halliday and Hasan's cohesive concepts and their categorization of cohesive devices, this paper analyzes 70 Chinese students' English writings, making the comparison between the estimation of coherence made by LSA and our experts. The results are consistent which indicate that LSA is proved to be an objective and appropriate method in measuring the textual coherence. Therefore, close attention should be paid to the cultivation of textual coherence, which will be of significance to the teaching of English writing.
|
PDF Full Text Request