**Sentiment Analysis of Movie Reviews**
Divya Bhaskaran, Akhilesh Reddy Baddigam
*University of Texas at San Antonio, San Antonio, Texas, USA*
{ioa034, qlb680}@my.utsa.edu
**Project Definition:** Convolution neural network is used in all the machine learning applications; it also eases the classification of a sentence whether it is a positive review or negative review by the reviews given by the viewers to a movie. In our model we use word embedding, which is used to reduce the dimension of the large vocabulary to low dimension space vectors as input.
Our approach towards the text classification problem is by labelling the dataset in to positive and negative sets, processing data in to required format, applying convolution filters, get a predicted output and training the data is to be carried out using different functions and techniques available in tensor flow. The main idea is to play with libraries available in tensor flow and increase the efficiency with that of the present one.
[1]: YoonKim, (Sep 2014). Convolutional Neural Networks for Sentence Classification. New York University
**Outcome:** Applying Convolutional Neural network for Sentence classification and improving the efficiency of the prediction.
**Dataset:** This dataset is taken from Cornell website. This site contains a number of dataset in specific we would like to use, sentence polarity dataset v1.0.movie review data set, which contains 5331 positive review and 5331 negative review as individual sentences.
In [ ]: