Deep Learning Based Twitter Sentiment Analysis

Abstract

To summarize reviews of a product, an extensively used approach is Aspect-based text mining which is applied on tweets or review data. With the help of Twitter sentiment detectors (TSDs) we can obtaina finer answer to assess a product in terms of quality and service. Social media has become a necessarymeans of communication in our daily lives. If we consider Twitter, most of the tweets are usually unstructured, non-formal, and it lacks labeled data like ratings, emojis, different categories, images, hashtags, videos etc. which makes it laborious for product opinion mining.