Penelitian ini menganalisis sentimen terhadap layanan Pendaftaran Mahasiswa Baru (PMB) ITPLN di Instagram menggunakan 579 komentar yang dikumpulkan selama periode 6 bulan dari Januari hingga Juni 2024. Proses text preprocessing melibatkan penghapusan data duplikat, cleaning, case folding, normalization, tokenization, dan stemming. Setelah text preprocessing, data diberi label menggunakan metode Lexicon Based dan dibobot dengan metode Term Frequency-Inverse Document Frequency (TF IDF). Untuk klasifikasi sentimen, penelitian ini menggunakan dua metode: Naïve Bayes dan Long-Short Term Memory (LSTM). Hasil menunjukkan bahwa metode LSTM, dengan akurasi 71%, lebih unggul dibandingkan Naïve Bayes yang memiliki akurasi 69%. Selain itu, hasil analisis sentimen menunjukkan bahwa komentar positif lebih dominan dibandingkan komentar negatif, mengindikasikan bahwa mayoritas pengguna Instagram memiliki pandangan positif terhadap layanan PMB ITPLN.

This study analyzes sentiment towards the New Student Enrollment (PMB) services of ITPLN on Instagram using 579 comments collected over a 6-month period from January to June 2024. The text preprocessing process involves removing duplicate data, cleaning, case folding, normalization, tokenization, and stemming. After text preprocessing, the data is labeled using the Lexicon-Based method and weighted using the Term Frequency Inverse Document Frequency (TF-IDF) method. For sentiment classification, this study uses two methods: Naïve Bayes and Long-Short Term Memory (LSTM). The results show that the LSTM method, with an accuracy of 71%, outperforms Naïve Bayes, which has an accuracy of 69%. Additionally, the sentiment analysis results indicate that positive comments are more dominant than negative comments, suggesting that the majority of Instagram users have a positive view of the PMB services at ITPLN.