DARNN: Discourse Analysis for Natural languages using RNN and LSTM
Madhuri A. Tayal, Animesh R. Tayal · INTERNATIONAL JOURNAL OF NEXT-GENERATION COMPUTING · 2021
Natural language processing (NLP) is a computer science arena concerned with computer connections and human (natural) language connections. One of the significant subfunctions of any application to the NLP is discourse processing. Discourse processing is intended to uncover information from certain information that is understood naturally (NLU). In fulfilling this task, we used special neural networks (RNN with LSTM). Recurrent neural networks are the prerequisites for (RNN). However, RNN does have some of the disadvantages of LSTM (Long Short-Term Memory). Kaggle and Keras are collected with data. The designed system proposed provides LSTM with 72% accuracy. With more text generation deep learning algorithms, the same approach can also be used.