Published Paper Details:

Sentiment analysis is the task of extracting contextual information from the text. The source of textual data can be from social media posts, comments, reviews, blogs, tweets, news articles etc...These text involves in stop word removal, lemmatization and stemming. After text preprocessing steps, the text has to processed using word embedding techniques. The word has to processed into vectors that can involve into machine learning and neural network approaches. Word2vec and skip gram are the two successful approaches in supervised machine learning and GLoVe in unsupervised machine learning. Recurrent Neural Network(RNN) and Convolution Neural Network (CNN) are the two approaches in deep learning algorithm. Thus the paper discusses the machine learning and deep learning algorithm in word embedding techniques can be used for feature extraction. Then it can be fed into various classifiers to fall under the category of positive, negative or neutral which helps to understand the moral and ethical values of people in various applications such as business, politics, ecommerce and health care industry and in movie reviews. The experimental results show that deep learning approaches produce better results in sentiment prediction when compared with Machine learning.