ANALISIS SENTIMEN REVIEW INSTITUSI PENDIDIKAN TINGGI DARI FACEBOOK MENGGUNAKAN JARINGAN SARAF TIRUAN

Habibullah Akbar

Sari


Abstract

Nowadays, the development of digital technology allows the higher institutions such as universities to observe public opinion/sentiment through social media. Unfortunately, the analysis of the data is still done manually. In this study, we develop an automated sentiment analysis framework to predict the polarity of public opinion on several higher education institutions in Indonesia. The data used comes from the comments that are available on the official page of the institution's Facebook. We make comparisons of several combined models between classifiers (Logistic Regression, Support Vector Machine, Random Forest Classifier and Neural Network) with feature extraction methods (Count vectorizer and TF-IDF). The experimental results show that the best sentiment analysis results are given by the Neural Network model which uses the TF-IDF weighting technique with an accuracy of more than 86%. These results indicate that the proposed sentiment analysis framework is potential to be used to find out public opinion on higher education institutions in Indonesia.

 

Keywords: Data mining, sentiment analysis, machine learning.

 

Abstrak

Di masa sekarang ini, perkembangan teknologi digital memungkinkan institusi pendidikan tinggi seperti universitas untuk mengetahui opini/sentimen masyarakat lewat media sosial. Sayangnya, analisis terhadap data tersebut masih dilakukan secara manual. Pada penelitian ini, kami membangun sebuah kerangka analisis sentimen otomatis untuk memprediksi polaritas dari opini masyarakat terhadap beberapa institusi pendidikan tinggi di Indonesia. Data yang digunakan berasal dari komentar-komentar yang berada pada halaman resmi Facebook institusi yang digunakan untuk penelitian. Kami melakukan perbandingan dari beberapa model gabungan antara klasifikator (Logistic Regression, Support Vector Machine, Random Forest Classifier dan Neural Network) dengan metode ekstraksi fitur (Count vectorizer dan TF-IDF). Hasil eksperimen menunjukkan bahwasanya hasil analisis sentimen terbaik diberikan oleh model Neural Network yang menggunakan teknik pembobotan TF-IDF dengan akurasi lebih dari 86%. Hasil ini menunjukkan bahwa kerangka analisis sentimen yang diusulkan berpotensi untuk digunakan untuk mengetahui opini masyarakat terhadap institusi pendidikan tinggi yang ada di Indonesia.   

 

Kata kunci: Data mining, analisis sentimen, machine learning. 


Teks Lengkap:

PDF

Referensi


Abbasi, A., Chen, H., & Salem, A. (2008). Sentiment analysis in multiple languages: Feature selection for opinion classification in web forums. ACM Transactions on Information Systems (TOIS), 26(3), 12.

Abdelrazeq, A., Janßen, D., Tummel, C., Jeschke, S., & Richert, A. (2016). Sentiment analysis of social media for evaluating universities. In Automation, Communication and Cybernetics in Science and Engineering 2015/2016 (pp. 233–251). Springer.

Akbar, H., Suryana, N., & Sahib, S. (2011). Training neural networks using Clonal Selection Algorithm and Particle Swarm Optimization: A comparisons for 3D object recognition. In 2011 11th International conference on hybrid intelligent systems (HIS) (pp. 692–697).

Altrabsheh, N., Gaber, M. M., & Cocea, M. (2013). SA-E: sentiment analysis for education. In International Conference on Intelligent Decision Technologies (Vol. 255, pp. 353–362).

Cummins, S., Burd, L., & Hatch, A. (2010). Using feedback tags and sentiment analysis to generate sharable learning resources investigating automated sentiment analysis of feedback tags in a programming course. In 2010 10th IEEE International Conference on Advanced Learning Technologies (pp. 653–657).

Hamdan, H., Bellot, P., & Bechet, F. (2015). Lsislif: Crf and logistic regression for opinion target extraction and sentiment polarity analysis. In Proceedings of the 9th international workshop on semantic evaluation (SemEval 2015) (pp. 753–758).

Kechaou, Z., Ammar, M. Ben, & Alimi, A. M. (2011). Improving e-learning with sentiment analysis of users’ opinions. In 2011 IEEE global engineering education conference (EDUCON) (pp. 1032–1038).

Knerr, S., Personnaz, L., & Dreyfus, G. (1990). Single-layer learning revisited: a stepwise procedure for building and training a neural network. In Neurocomputing (pp. 41–50). Springer.

Koehler, M., Greenhalgh, S., & Zellner, A. (2015). Potential Applications of Sentiment Analysis in Educational Research and Practice--Is SITE the Friendliest Conference? In Society for Information Technology & Teacher Education International Conference (pp. 1348–1354).

Mollett, A., Moran, D., & Dunleavy, P. (2011). Using Twitter in university research, teaching and impact activities. LSE Public Policy Group, London School of Economics and Political Science.

Wen, M., Yang, D., & Rose, C. (2014). Sentiment Analysis in MOOC Discussion Forums: What does it tell us? In Educational data mining 2014.


Refbacks

  • Saat ini tidak ada refbacks.


VISITOR COUNTER:

gerEGGe