A Dual Attention Encoder-Decoder Text Summarization Model

A worthy text summarization should represent the fundamental content of document. Recent studies on computerized tried to present solutions this challenging problem. Attention models are employed extensively in process. Classical attention techniques utilized acquire context data decoding phase. Nevertheless, without real and efficient feature extraction, produced summary may diverge from core topic. In article, we an encoder-decoder system employing dual mechanism. mechanism, algorithm gathers main encoder side. model, can capture produce more rational content. The merging two phases produces precise summaries. enhanced mechanism gives high score repetition increase phrase score. It also captures relationship between phrases title giving them higher We assessed our proposed model with or significance optimization using ablation procedure. Our achieved highest performance 96.7% precision least CPU time among other both training sentence extraction.