• Joshua Fernandes
• Eldridge Fernandes
• Ronak Sakhiya
• Sourabh Mayekar
• Prathibha Sudhakaran

Sentiment Analysis is the analysis of the sentiment of the sentence based on the type of the word used. In this digital era, there are huge datasets of reviews available from customers across the globe. It is tedious to analyze the large datasets of reviews manually one by one. In this paper, we proposed to use a bunch of models for the analysis of sentiments. The models we employed are Support Vector Machine (SVM), Logistic Regression (LR), Multilevel Perceptron (MLP), Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM) Network. We used the standard IMDB datasets for all these five models. IMDB dataset is one of the most reputable dataset and its well-labeled data makes it easy to carry out operations which you intend to quiet feasibly. These datasets consist of reviews from different movies. Using this dataset, we tried to compare the accuracy obtained by each of the models. As per our results obtained we can certainly state that deep learning models are able to outperform the other models as they are able to analyze a deeper relationship within the sentence which drastically improves its classification with every epoch.

Sentiments analysis, IMDB dataset, SVM, LR, MLP, CNN, and LSTM.

"Sentiment Analysis On Text Using Neural Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.420-426, April-2019, Available :http://www.jetir.org/papers/JETIR1904P67.pdf

"Sentiment Analysis On Text Using Neural Network", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp420-426, April-2019, Available at : http://www.jetir.org/papers/JETIR1904P67.pdf