A Deep Learning Approaches for Natural Language Processing and Sentiment Analysis in Social Media
Ashiq VM, E. J. Thomson Fredrik, N. Krishna Kumar, Fred Torres‐Cruz, Jordan Piero Borda Colque, Geetha Manoharan
Abstract
Twitter and Facebook are far more effective at disseminating information than other, more established social media platforms. As social media has grown, it has developed into a gold mine of information that companies as well as scholars can use to create models to analyze this repository and acquire data that can be used for tactics involving word-of-mouth advertising. For these rapid information exchanges, many Natural Language Processing (NLP) techniques that concentrate on identifying formal terms are inappropriate. On the other hand, the vocabulary used on social media is somewhat specialized and contains certain symbols. In this work, we present a novel deep learning-based methodology for social media sentiment analysis. From the data we collect, we build a dataset. After analysing these specific phrases, our goal is to produce a semantic dataset to aid future research. Retrieved data are greatly useful for future applications. Multiple platforms of social media are crawled using Python for gathering trial data.