Twitter Sentiment Analysis using Deep Learning
Ghazi A1, Fatih Ö2
1Ghazi A*, Software Engineering, Firat University, Elazig, Turkey.
2Fatih Ö, Assistant Professor, Software Engineering, Firat University, Elazig, Turkey.
Manuscript received on April 20, 2020. | Revised Manuscript received on May 01, 2020. | Manuscript published on May 10, 2020. | PP: 1040-1044 | Volume-9 Issue-7, May 2020. | Retrieval Number: G5814059720/2020©BEIESP | DOI: 10.35940/ijitee.G5814.059720
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The whole world is changing rapidly with current innovations, using the Internet, has become a fundamental requirement in people’s lives. Nowadays, a massive amount of data made by social networks based on daily user activities. Gathering and analyzing people’s opinions are crucial for business applications when they are extracted and analyzed accurately. This data helps the corporations to improve product quality and provide better customer service. But manually analyzing opinions is an impossible task because the content is unorganized. For this reason, we applied sentiment analysis that is the process of extracting and analyzing the unorganized data automatically. The primary steps to perform sentiment analysis include data collection, pre-processing, word embedding, sentiment detection, and classification using deep learning techniques. This work focused on the binary classification of sentiments for three product reviews of fast-food restaurants. Twitter is chosen as the source of data to perform analysis. All tweets were collected automatically by using Tweepy. The experimented dataset divided into half of the positive and half of the negative tweets. In this paper, three deep learning techniques implemented, which are Convolutional Neural Network (CNN), Bi-Directional Long Short-Term Memory (Bi-LSTM), and CNN-Bi-LSTM, The performance of each of them measured and compared in terms of accuracy, precision, recall, and F1 score Finally, Bi-LSTM scored the highest performance in all metrics compared to the two other techniques.
Keywords: Sentiment Analysis, CNN, Bi-LSTM, NLP (Natural Language Processing).
Scope of the Article: Deep Learning