The Research On Corpus-based Computer-Assisted Instruction Of English

Corpora are banks of electric texts that are collected from texts or speeches by means of random sampling according to certain linguistic principles. Corpora can provide learners with rich contexts in English learning, thus they are helpful to the construction of linguistic knowledge of English learners. Great development has taken place in the research of corpus in recent years, yet such a powerful tool as corpus has not been used widely in English teaching mostly because of its technological difficulties. For a long time corpus is mostly used in the studies of linguistics, the compilation of dictionaries and some important examinations but seldom in computer-assisted instruction of English. The potential of the application of corpora is tremendous. In the field of educational technology, almost no attention has been paid to the studies of corpus-based computer-assisted instruction of English.With the continuous increase of qualities and decrease of prices of computers, the proliferation of English electronic texts and the popularity of optical scanners in teaching, it is possible for English teachers and small English teaching teams to build their own corpora for English teaching. The application of corpora in English teaching is helpful to the change of the traditional teaching model which is characterized by taking teachers as centers of teaching. It can also be helpful to the application of data-driven learning which is based on constructivism. So we can say that the application of corpora in English teaching is of great significance from perspectives of both the construction of teaching resources and the reform of teaching.Firstly, this thesis introduces the concept, history, the frequently-used parameters and the methods of statistics and analyses of corpus as well as the software used in the studies of corpus. An outline of the basic knowledge and methods is elaborated.Secondly, the thesis elaborates constructivism and data-driven learning which are the theoretical bases of corpus-based computer-assisted instruction of English. After that the thesis elaborates the design of the learning environment based on constructivism. It also analyses an example of learning environment model based on constructivism that was given by Jonassen. The thesis gives the application of the model given by Jonassen in the field of corpus-based computer-assisted English teaching.Last but not least, in chapter four, the thesis discusses the classification of corpora for teaching, the resources of corpora and several examples of the application of corpus in computer-assisted English teaching. Among these examples the movie caption corpus is the most important creation in this thesis. The combination of English movies and movie caption corpora can provide the learners with double lively contexts, one is the voices and pictures in the movies, the other is the context composed of the movie captions. It is easy for learners to find a certain word or words that have some common features in corpora. After the analysis of the structure of the movie caption files (*.srt), the thesis design a tool to delete the texts that are not useful for the construction of corpora. With the help of this tool and the srt files on the Internet, a corpus that is composed of 1,100,000 tokens is constructed. The thesis also gives some examples of how to use this corpus.