An Improved Framework for Sentiment Analysis for College Reviews
Sri Devi T · International Journal of Advanced Trends in Computer Science and Engineering · 2020
This is the data era and large amount of data is processed every second for various purposes and deep learning is used to abstract the key aspects of data.Lack of core knowledge on deep learning makes this a Black Box machine which remains dark to the real world.In the recent years many researches are going in the field of Machine learning in which Deep Learning has gained more attention.The aim of the paper is to provide survey of trending deep learning algorithms.The algorithms such as Deep Neural Network (DNN), Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) are discussed in detail.Then, performance of RNN-LSTM algorithm is compared by training and predicting college reviews feedback of data set size 500 with epochs 1 to 7. It is identified that accuracy of RNN-LSTM algorithm was around 83%.This paper provides better understanding of the deep learning algorithms and their implementation in various research areas.