Ijraset Journal For Research in Applied Science and Engineering Technology
Authors: Sheetal J, Mahesh Thotar, Veerendra Kumar K, Yuva Sai M, S Uttej Kumar M
DOI Link: https://doi.org/10.22214/ijraset.2024.62060
Certificate: View Certificate
This project presents an intelligent system political sentiment analysis with information from Twitter. . The system utilizes deep learning techniques, specifically a BiLSTM model featuring an attention mechanism, to predict the political affiliation (BJP or Congress) of tweets. The data preprocessing involves cleaning text, data augmentation, tokenization, and indexing. A dataset is used to train the model. consisting of tweets from both parties, getting very accurate results in predicting the sentiment of tweets. Additionally, the software offers functionality for data visualization, including displaying sample tweets, sentiment analysis results, and the winning party. Additionally, an admin page is available for further customization and management. Overall, this project the possibilities of Using deep learning to analyze sentiment of political content on social media data.
I. INTRODUCTION
Social media platforms like Twitter have become crucial sources of communication and data for political parties and their supporters. Understanding public sentiment on these platforms is essential for political parties to gauge their popularity, assess public opinion, and strategize their campaigns effectively. In this context, Utilizing Twitter data for political sentiment analysis has drawn a lot of interest in recent years.
This project focuses on analyzing the political sentiment expressed in representing the two largest political parties in India: the Bharatiya Janata Party (BJP) and the Indian National Congress (Congress). By collecting and analyzing tweets from these parties, we aim to learn more about how the public feels about each party and their policies.
The project includes several steps to achieve its objectives. First, we plot sample tweets from both parties to showcase the diversity of content and sentiment expressed on Twitter. Then, we generate a pie diagram to visually represent the distribution of tweets between the BJP and Congress. Next, we construct histograms to illustrate the sentiment distributions for each party, an enhanced comprehension of the sentiment conveyed through the tweets.
To further explore the content of the tweets, we create a word cloud highlighting the most frequent terms used in the tweets. This visualization helps identify key themes and topics of discussion surrounding each political party.
To predict the sentiment of the tweets, we employ a Bidirectional Long Short-Term Memory deep learning model (BiLSTM) with an attention mechanism. This model is trained to classify tweets as either positive or negative according to their content. The predictions from the model are then aggregated to determine the overall sentiment for each political party.
Additionally, we provide an accuracy plot in order to assess the efficiency of our sentiment analysis model. Finally, an admin page is included for further customization and management of the application.
Through this project, we aim give insightful information about public sentiment towards the BJP and Congress on Twitter, using cutting-edge methods for data analysis and deep learning models.
II. LITERATURE SURVEY
III. METHODOLOGY
2. Data Visualization:
3. Word Cloud Generation:
4. Sentiment Analysis Model:
5. Model Training and Evaluation:
6. Sentiment Prediction and Analysis:
7. Admin Page Implementation:
8. Deployment and Presentation:
A. Implementation
Plot Text Data:
2. Pie Diagram:
3. Histogram of Sentiments:
4. Word Cloud:
5. Accuracy Plot:
6. Prediction:
7. Admin Page:
V. MACHINE LEARNING ALGORITHMS APPLIED
2. Attention Mechanism:
3. Tokenization and Embedding:
4. Cross-Entropy Loss:
5. Adam Optimizer:
In this study, Using Twitter data, we showed how effective deep learning approaches are for forecasting election outcomes. By leveraging the vast amount of real-time information available on Twitter, our model was able to accurately forecast election outcomes. The integration of sentiment analysis and NLP allowed us to capture nuanced public sentiment, contributing to the predictive power of our model. As social media continues to play a significant role in shaping public opinion, Our method yields insightful information for political analysts and policymakers. Additionally, our findings underscore the potential of deep learning methodologies in harnessing big data for sociopolitical forecasting. Further research could explore the implimentation of similar techniques to different electoral contexts and investigate the robustness of our model across diverse demographic and cultural landscapes. Ultimately, our work contributes to advancing the field for computational social science and offers a promising avenue for improving the correctness and timeliness of election predictions.
[1] H. Varian & H. Choi (2012). use Google Trends to forecast the present. Record of Economics, 88(s1),2–9. [2] David Gayo-Avello (2012). No, Twitter is not a reliable way to predict elections. 91–94 in IEEE Internet Computing, 16(6). [3] Schoen, H., Jurgens, P., and Jungherr, A. (2012). A rebuttal to Tumasjan, A., Sprenger, T. O., Sandner, P. G., & Welpe, I. M. (2011) on why the pirate party won the 2009 German election or the problem with forecasts. 30(2), 229–234 in Social Science Computer Review. [4] Sandner, P. G., Welpe, I. M., Sprenger, T. O., and Tumasjan, A. (2011). Predicting elections with Twitter: What 140 characters convey about political mood. 11(1) ICWSM, 178-185. [5] In 2013 Gayo-Avello, D. A meta-analysis of the most advanced election forecast using data from Twitter. 31(6), 649–679, Social Science Computer Review. [6] Zeng, X., Mao, H., and Bollen, J. (2011). Stock market sentiment can be predicted by Twitter mood. 2(1), 1-8; Journal of Computational Science. [7] DiGrazia, J., Bollen, J., McKelvey, K., and Rojas, F. (2013). Increased tweets and votes: Social media as a numerical measure of political conduct. 8(11), e79449. PloS one. [8] Bond (2012) cited Fariss, C. J., Jones, J. J., Kramer (2012), Marlow (2012), Settle (2012), and Bond (2013). a 61-million-person social influence and political mobilization exercise. 295-298 in Nature, 489(7415). [9] Barberá (2015) p. Combine tweets from similar individuals to estimate Bayesian ideal points using Twitter data. 76–91 in Political Analysis, 23(1). [10] J. Grimmer & B. M. Stewart (2013). Text as data: The benefits and drawbacks of using automated content analysis techniques on political texts.
Copyright © 2024 Sheetal J, Mahesh Thotar, Veerendra Kumar K, Yuva Sai M, S Uttej Kumar M. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Paper Id : IJRASET62060
Publish Date : 2024-05-13
ISSN : 2321-9653
Publisher Name : IJRASET
DOI Link : Click Here