Analisis Sentimen dan Pemodelan Topik pada Ulasan Pengguna Aplikasi myIM3 Menggunakan Support Vector Machine dan Latent Dirichlet Allocation
Abstract
In the current digital era, mobile applications play a crucial role in enhancing user experience. This study analyzes user sentiment towards the myIM3 application and identifies key topics discussed in user reviews using Support Vector Machine (SVM) and Latent Dirichlet Allocation (LDA). The dataset comprises 1,000 user reviews from the Google Play Store, including review text, star ratings, review dates, and application versions. Data preprocessing involved cleaning, normalization, stop word removal, and lemmatization. Text data was transformed using Term Frequency-Inverse Document Frequency (TF-IDF). The dataset was split into training and testing sets (80:20 ratio). The SVM model, optimized with a linear kernel, achieved an accuracy of 84.65%, with a precision of 85% for negative sentiment, 84% for positive sentiment, and challenges in classifying neutral sentiment. Cross-validation ensured model robustness. LDA identified five primary topics: general user experience, application usability and purchase experience, positive feedback and functionality, general application evaluation, and network issues and pricing concerns. Techniques like oversampling, undersampling, and hybrid methods addressed imbalanced datasets to enhance model performance. The results revealed that 43% of reviews were positive, 42% were negative, and 15% were neutral. The key topics indicated that network issues and pricing were significant user concerns. These findings provide valuable insights for developers and stakeholders to improve user experience and refine application features based on user feedback.
Downloads
References
J. Huang, "Analysis of Airline Tweets Using Support Vector Machines," Journal of Transportation Research, vol. 22, no. 3, pp. 45-56, 2023.
Cahyo, A., Nugroho, B., and Sudirman, T., "Sentiment Analysis of IMDb Movie Reviews Using TF-IDF and SVM," Journal of Data Science, vol. 15, no. 1, pp. 34-42, 2023.
Benarafa, A., Rahman, F., and Lee, S., "Addressing Overfitting and Underfitting in Sentiment Analysis with WordNet Integration," Journal of Computational Linguistics, vol. 28, no. 2, pp. 123-135, 2023.
R. Hokijuliandy, Pratama, D., and Aditya, M., "Improving Sentiment Analysis of Mobile JKN Application Reviews Using Chi-Square Feature Selection and SVM," Journal of Health Informatics, vol. 10, no. 4, pp. 89-98, 2023.
R. Driyani, "Sentiment Analysis of Mobile Reviews on Twitter Using SVM with RBF Kernel," Journal of Social Media Studies, vol. 12, no. 2, pp. 200-215, 2021.
N. Obiedat, Alharbi, K., and Zhang, L., "Hybrid Approaches to Addressing Imbalanced Datasets in Sentiment Analysis," Journal of Artificial Intelligence Research, vol. 35, no. 5, pp. 67-78, 2022.
Putra, A., Wibowo, H., and Kusuma, F., "Enhancing Sentiment Analysis with SentiWordNet and SVM: A Hybrid Approach," Journal of Language Technology, vol. 18, no. 3, pp. 155-169, 2021.
Smith, D., "Machine Learning Techniques for Sentiment Analysis," International Journal of Data Science, vol. 20, no. 4, pp. 110-120, 2022.
Jones, T., "Text Mining and Sentiment Analysis with SVM and LDA," Journal of Information Technology, vol. 14, no. 2, pp. 85-95, 2021.
Lee, H., "Improving Sentiment Classification with Hybrid Models," Journal of Computational Intelligence, vol. 25, no. 3, pp. 99-108, 2022.
Williams, K., "Sentiment Analysis Using Advanced Machine Learning Techniques," Journal of Artificial Intelligence and Robotics, vol. 27, no. 1, pp. 60-70, 2023.
Brown, P., "Evaluating the Effectiveness of Sentiment Analysis Models," Journal of Data Mining and Knowledge Discovery, vol. 30, no. 2, pp. 130-140, 2021.
Kim, S., "Challenges and Solutions in Sentiment Analysis," Journal of Machine Learning Research, vol. 22, no. 5, pp. 145-155, 2023.
Ali, M., "A Comparative Study of Machine Learning Algorithms for Sentiment Analysis," Journal of Big Data Analytics, vol. 17, no. 3, pp. 75-85, 2022.
Patel, R., "Using SVM and LDA for Sentiment Analysis in Social Media," Journal of Social Network Analysis, vol. 19, no. 4, pp. 205-215, 2021.
Green, J., "Web Scraping Techniques for Data Collection in Sentiment Analysis," Journal of Data Engineering, vol. 10, no. 2, pp. 89-102, 2022.
Zhang, Y., "Cross-Validation Techniques for Machine Learning Models," Journal of Computational Statistics, vol. 28, no. 3, pp. 110-122, 2023.
Lin, J., "Latent Dirichlet Allocation for Topic Modeling in Reviews," Journal of Text Analytics, vol. 15, no. 1, pp. 45-60, 2023.
Zhao, F., "Evaluating Coherence in Topic Modeling with LDA," Journal of Machine Learning Techniques, vol. 18, no. 4, pp. 75-88, 2021.
Torres, M., "Handling Imbalanced Datasets in Sentiment Analysis," Journal of Advanced Data Analytics, vol. 14, no. 2, pp. 50-65, 2022.
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Analisis Sentimen dan Pemodelan Topik pada Ulasan Pengguna Aplikasi myIM3 Menggunakan Support Vector Machine dan Latent Dirichlet Allocation
Pages: 1618-1626
Copyright (c) 2024 Priyo Agung Prastyo, Berlilana Berlilana, Imam Tahyudin

This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).