Text Automatic Summarization Generation Algorithm for English Teaching

During the process of English teaching, the teachers should have the ability to extract the English text summarization rapidly and analyze certain type of information comprehensively. While in the process of summarization automatic generation, perfect similarity computation plays important role for successful summarization. A novel similarity computation method is proposed in this paper, which improves traditional LDA method. It marks the word in text as three types and carries on LDA modeling respectively according to the word sets of different parts of speech. Then, integrated with this method, an English text automatic summarization scheme is further put forward. The experimental results show that the improve scheme is superior to other similar algorithms in every ROUGE evaluation metrics, and it has advantage compared to other LDA-based summarization algorithms.