Sentiment Emotions Scoring for Apple Mobile Tweets

Abstract

Sentimental analysis is one of the hottest topic on social media and used in web applications. Sentiment analysis is all about expressing their emotions and opinions on a particular topic or   on the particular product in the form of reviews and tweets. Sentiment analysis can express their emotions in the form of positive, negative or in the form of neutral.  The product performance or product usage can be known through sentimental analysis. In our proposed system, the dataset chosen for sentiment analysis is apple mobile phone dataset one is before earnings and the other one is after the earnings. The datasets is applied to pre-processing strategies to remove inconsistent and redundant factors. The proposed methods of pre-processing include the deletion of punctuations, special characters, numbers, HTML characters escaping. The dataset is further fine tuned by applying decoding data, Apostrophe Lookup, removing stop words, removing URLs and eliminating expressions. Visualization of the preprocessed information is represented as word cloud with the frequency facts of the key words. Finally the tweets are classified into emotions based on nrc_sentiment dictinory and descriptive analysis for the emotions.