Sentiment Analysis Using Long Short-Term Memory on Twitter Dataset

Le résumé fourni par la source

In the new technological era of social media, a sentiment analysis grown as a pivotal and important tool for getting public opinion and sentiment expressed in the form of textual data. There many social media platforms, among these a twitter as a prominent social media platform, allows a vast dataset of user generated content. This makes a compelling source for sentiment analysis research. This research article addresses the application of Long Short Term Memory (LSTM), which is specialized form Recurrent Neural Network (RNN). The main aim of the research is to encompass the development of novel LSMT based technique for accurately sentiment analysis and exploring of trends of sentiment trends on Twitter dataset. We propose a LSMT model architure to address the dynamic nature on Twitter dataset based on by capturing inherent capacity to sequential dependencies. For ensuring effectiveness of model, we delve into pre-processed method, feature engineering. The observed performance shows that the LSTM-based technique higher accuracy and efficacy as compared to traditional sentiment analysis. Furthermore, our analysis shows the nuanced dynamic shifts in sentiment and sentiment patterns, helpful valuable insights in the field of emerging field of social monitoring and research. Our research work enhances the understanding of analysis of sentiment opinion techniques and shows the capabilities of LSTM networks for capturing sentiment dynamics in the rapidly growing social media platform that demonstrating the interplay of opinions and emotions within the Twitter background.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.