Automatic-Text-Generation

Artificial Intelligence, Neural Network and deep learning appearance brought innovation in an individual stream like finance, music, games, medicine, telecommunication, literature and other fields. Neural networks and deep learning are potent environments to solve the problems related to sequencing. In recent years the deep learning proves the notable work on the generation of sequence problem. Generation of text or sequence in a meaningful way without losing its importance is challenging exercise and. Various efforts have been applied over the time with the diverse results. Deep learning development already has multiple models for the generation of sequence text, but LSTM and RNN provide the best optimization on the sequencing problem. This paper aims at automatic generation of sequence based on the inputted story text by using long short-term memories (LSTM) and recurrent neural network (RNN). The designed network generates the outcome in the similar context of input sequence with different thresholds. In addition to that, it identifies the best optimization and error loss comparison of long short-term memories and recurrent neural network at similar and dissimilar epochs.