Research and Development of Network Multimedia English Language Teaching Model Based on Corpus

Modern corpus is the database of real language electronic texts. Users can extract all the sentences containing keywords or structures from the corpus through the retrieval program, make structural comparison and word analysis, and summarize language phenomena. The combination of corpus and multimedia provides a huge information database for teaching. Because of its convenient access, it greatly reduces the time and workload of educators in preparing lessons, and can also promote learners’ autonomous learning. This paper presents an assessment model of English language teaching (ELT) based on weighted reasoning model, and its effectiveness is verified by comparison with other assessment algorithms. Then, this assessment model is used to analyze the significance of corpus in the construction of network multimedia ELT model. The results show that, compared with the traditional back propagation neural network (BPNN), the proposed teaching assessment model has higher prediction accuracy and recall rate, which are increased by $19.69 \%$ and $16.88 \%$ respectively. Therefore, this method is feasible for ELT assessment. From the assessment of instructional effect, it can be seen that the application of corpus in the innovation of network multimedia ELT has a remarkable effect.