ANALISIS BRAND LAYANAN AKADEMIK PERGURUAN TINGGI INDONESIA MENGGUNAKAN KLASIFIKASI TEKS DI MEDIA SOSIAL

Hashri Hayati; Muhammad Riza Alifi

doi:10.46576/syntax.v6i1.6924

ANALISIS BRAND LAYANAN AKADEMIK PERGURUAN TINGGI INDONESIA MENGGUNAKAN KLASIFIKASI TEKS DI MEDIA SOSIAL

Hashri Hayati, Muhammad Riza Alifi

Abstract

Penelitian ini bertujuan untuk menganalisis persepsi komunitas eksternal terhadap brand akademik perguruan tinggi di Indonesia melalui media sosial, khususnya Twitter/X. Seiring dengan tingginya jumlah perguruan tinggi dan angka partisipasi kasar (APK), kompetisi antar institusi pendidikan tinggi semakin kuat, mendorong perlunya diferensiasi brand yang disampaikan ke publik. Dalam studi ini, dikumpulkan post dari 30 akun resmi X perguruan tinggi di Indonesia yang kemudian diklasifikasikan ke dalam lima kategori brand akademik: Innovative, Global Impact, Student Engagement, Career Focused, dan Research Excellent. Proses klasifikasi dilakukan dengan membangun model pembelajaran menggunakan algoritma Naïve Bayes, yang diimplementasikan melalui pustaka pemrosesan bahasa alami di lingkungan Node.js. Untuk mengevaluasi kinerja model, dilakukan pengujian terhadap dataset uji terpisah, dan dihitung metrik evaluasi berupa precision, recall, dan accuracy berdasarkan nilai True Positive, False Positive, dan False Negative yang diperoleh melalui confusion matrix untuk setiap kelas. Hasil evaluasi menunjukkan bahwa model yang dikembangkan memiliki performa nilai rata-rata precision sebesar 80,8%, recall sebesar 78,8%, dan accuracy sebesar 80%, sehingga dapat diandalkan sebagai alat bantu untuk memahami kesesuaian antara brand yang dikomunikasikan dan persepsi publik secara daring.

Kata Kunci— brand akademik, brand perguruan tinggi, klasifikasi teks, naïve bayes, media sosial.

ABSTRACT

This study aims to analyze the perceptions of external communities regarding the academic branding of Indonesian universities through social media, particularly Twitter/X. With the growing number of higher education institutions and rising gross enrollment rates, competition among universities has intensified—prompting the need for more distinct and strategic public brand positioning. In this study, posts were collected from 30 official university X accounts in Indonesia and categorized into five academic brand themes: Innovative, Global Impact, Student Engagement, Career Focused, and Research Excellent. The classification process involved building a supervised machine learning model using the Naïve Bayes algorithm, implemented with a natural language processing library in the Node.js environment. To evaluate the model's performance, a separate test dataset was used, and evaluation metrics—namely precision, recall, and accuracy—were calculated for each class based on values of True Positive, False Positive, and False Negative derived from a confusion matrix. The results indicate that the developed model performs well, achieving average scores of 80,8% for precision, 78,8% for recall, and 80% for accuracy, making it a reliable tool for assessing the alignment between institutional brand communication and public perception in online discourse.

Keywords—academic brand, university brand, text classification, naïve bayes, social media.

Full Text:

PDF (Bahasa Indonesia)

References

M. F. Rouf, A. N. R. Attamimi, D. A. V. Putri, I. Nirmala, and A. N. Fadhilah, Statistik Pendidikan Tinggi. Direktorat Jenderal Pendidikan Tinggi, Riset, dan Teknologi (Kemendikti), 2023.

A. Wæraas and M. N. Solbakk, ‘Defining the essence of a university: lessons from higher education branding’, High Educ (Dordr), vol. 57, no. 4, pp. 449–462, 2009, doi: 10.1007/s10734-008-9155-z.

R. J. Brodie, M. S. Glynn, and V. Little, ‘The service brand and the service-dominant logic: missing fundamental premise or the need for stronger theory?’, Marketing Theory, vol. 6, no. 3, pp. 363–379, 2006, doi: 10.1177/1470593106066797.

J. Pringle and R. Naidoo, ‘Branding and the commodification of academic labour. In R. Barnett, P. Temple, & P. Scott (Eds.), Valuing higher education: An appreciation for the work of Gareth Williams’, UCL Institute of Education Press, pp. 158–177, 2016.

R. Rutter, S. Roper, and F. Lettice, ‘Social media interaction, the university brand and recruitment performance’, J Bus Res, vol. 69, no. 8, pp. 3096–3104, Aug. 2016, doi: 10.1016/j.jbusres.2016.01.025.

A. Peruta and A. B. Shields, ‘Social media in higher education: understanding how colleges and universities use Facebook’, Journal of Marketing for Higher Education, vol. 27, no. 1, pp. 131–143, Jan. 2017, doi: 10.1080/08841241.2016.1212451.

M. Z. Rabbil, T. D. Gugat, R. J. Intiha, and D. H. Putri, ‘Strategi Media Sosial yang Efektif untuk Meningkatkan Brand Awareness dan Engagement pada Kampus Politeknik Bina Madani’, Masarin, vol. 1, no. 2, pp. 67–77, Dec. 2022.

M. Yusuf, H. Saleh, and L. Setiawan, ‘Pengaruh Media Sosial Dan Citra Institusi Terhadap Minat Pada Keputusan Memilih Kuliah Pada Jurusan Perjalanan Politeknik Pariwisata Makassar’, Indonesian Journal of Business and Management, vol. 7, no. 1, pp. 152–159, 2024.

T. P. Yazid, A. Rasyid, and M. Hatika, ‘Pengaruh Cyber Public Relations terhadap Citra Perguruan Tinggi Swasta Terfavorit di Provinsi Riau’, Edukatif: Jurnal Ilmu Pendidikan, vol. 4, no. 4, pp. 5734–5743, 2022.

M. Beverland, Building brand authenticity, 1st ed. Basingstoke, England: Palgrave Macmillan, 2009.

Z. Wang, X. Sun, D. Zhang, and X. Li, ‘An Optimal SVM-Based Text Classification Algorithm’, in 2006 International Conference on Machine Learning and Cybernetics, 2006, pp. 1378–1381. doi: 10.1109/ICMLC.2006.258708.

J. Han, M. Kamber, and J. Pei, Data Mining: Concepts and Techniques. in The Morgan Kaufmann Series in Data Management Systems. Morgan Kaufmann, 2011. [Online]. Available: https://books.google.co.id/books?id=pQws07tdpjoC

J. Pringle and S. Fritz, ‘The university brand and social media: using data analytics to assess brand authenticity’, Journal of Marketing for Higher Education, vol. 29, no. 1, pp. 19–44, Jan. 2019, doi: 10.1080/08841241.2018.1486345.

M. Junker, R. Hoch, and A. Dengel, ‘On the evaluation of document analysis components by recall, precision, and accuracy’, in Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR ’99 (Cat. No.PR00318), 1999, pp. 713–716. doi: 10.1109/ICDAR.1999.791887.

DOI: https://doi.org/10.46576/syntax.v6i1.6924