Publisher

Tel-UTel-U

Journal International Journal on Information and Communication Technology (IJoICT)International Journal on Information and Communication Technology (IJoICT) Abstract

The evolution of social media has transformed platforms like Twitter from a mere information repository to a platform for expressing opinions and aspirations. Sentiment analysis on Twitter, particularly concerning the 2024 Indonesian presidential election, holds crucial importance for understanding public sentiment. The main contribution of this research is to optimize the Gated Recurrent Unit (GRU) model using Genetic Algorithm and combine feature expansion with Word2Vec for sentiment analysis on the topic of presidential election in Indonesia 2024. This research uses 39,791 datasets with GRU method, TF-IDF feature extraction, Word2Vec feature expansion with 142,545 corpus from IndoNews, and Genetic Algorithm optimization, the study achieves a peak accuracy of 86.46%, a 4.49% improvement over the baseline. By combining TF-IDF with a 5,000 maximum features, applying Word2Vec to the top 1 similarity, and utilizing Genetic Algorithm for feature optimization, this study demonstrates the effectiveness of these methods in improving the accuracy of sentiment analysis, thus significantly contributing to understanding public opinion during the 2024 Indonesian presidential election.

Conclusion

This research successfully analyzed public sentiment towards the 2024 Indonesian presidential election using Twitter data.The combination of GRU, TF-IDF, Word2Vec, and Genetic Algorithm proved effective in enhancing sentiment analysis accuracy, achieving a peak accuracy of 86.These findings provide valuable insights for political organizations, survey agencies, and media companies seeking to understand public opinion and inform their strategies.

Future Research

Further research should explore the application of this model to Twitter data from diverse topics beyond the presidential election to assess its generalizability. Investigating the impact of varying GRU model parameters on analysis results is crucial for optimizing performance. Additionally, exploring different word vector representations from various corpora could further enhance the models accuracy. Finally, comparing the performance of the Genetic Algorithm with other feature optimization methods would provide a comprehensive evaluation of its effectiveness. These future studies will strengthen the findings of this research and improve the effectiveness of the GRU model in analyzing sentiment from Twitter data, ultimately contributing to a deeper understanding of public opinion in Indonesia. The integration of these approaches will allow for a more nuanced and accurate assessment of public sentiment, providing valuable insights for various stakeholders.

References

1. Sentiment Analysis on Social Media Using Word2Vec and Gated Recurrent Unit (GRU) with Genetic Algorithm... socjs.telkomuniversity.ac.id/ojs/index.php/ijoict/article/view/903Sentiment Analysis on Social Media Using Word2Vec and Gated Recurrent Unit GRU with Genetic Algorithm socjs telkomuniversity ac ojs index php ijoict article view 903

Metadata

File size	565.35 KB
Pages	16
DMCA	Report