Design of English Teaching Assisted System using Tokenizer Attention Mechanism based Long Short-Term Memory

English assisted system has gained greater popularity and a part of national English education due to the demands of international economy as well as international politics. The important task in assisted system to improve educational quality and speed up educational modernization that increases and development for educational informatization. However, the teaching model in English assisted system has limitation of extracting limited keywords from specific texts which reduced the effectiveness of teaching system. To overcome this, a Tokenizer Attention Mechanism based Long Short-Term Memory (TAM-LSTM) is employed to enhance English assisted teaching system. Initially, system translation using corpus and rule based methods are described and utilized an improved chart parsing algorithm for translation of English texts. Then, translation memory format with basic entry and format entry is designed for storing the translated data. Finally, the proposed TAM-LSTM model in assisted teaching system enhanced teaching process of English by extracting important keywords effectively. The experimental results of proposed method achieved 85.82% accuracy when compared to existing method LSTM.