• T. TRITVA JYOTHI KIRAN

A Recurrent Neural Network (RNN) is a type of Artificial Neural Networks (ANN) that is designed to take temporal dimension into consideration by having a memory (internal state) (feedback loop). LSTM networks work better compared to RNN since they overcome the vanishing gradient problems. In practice, RNN fails to establish long-term dependencies Feed Forward Neural Networks (vanilla networks) that map a fixed size input (such as an image) to a fixed size output (classes or probabilities). But a drawback in Feed-Forward Networks is that they do not have any time dependency or memory effect. RNN allows us to work with a sequence of vectors: Sequence in inputs, Sequence in outputs, Sequence in both. In this work, I have used the Text dataset which contains over 1115394 characters, from which sequence of different combinations of unique characters and required selected sentences extracted from the text dataset and also make the effect of Dictionary. All the results are very efficient and accurate. The LSTM RNN proved its efficiency of learning 100% on Text Classification and Multi-Task Learning in my Performance analysis work.

RNN (Recurrent Neural Networks), LSTM (Long Short-Term Memory), Feed Forward Neural Networks (FF-NN), Text Classification, Gradient descent, loss.

"Text Predictions of LSTM RNN Performance Analysis using Tensor Flow GPU – Deep Adaptive Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 5, page no.b1-b9, May 2021, Available :http://www.jetir.org/papers/JETIR2105130.pdf

"Text Predictions of LSTM RNN Performance Analysis using Tensor Flow GPU – Deep Adaptive Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.8, Issue 5, page no. ppb1-b9, May 2021, Available at : http://www.jetir.org/papers/JETIR2105130.pdf