Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : Algoritme Jurnal Mahasiswa Teknik Informatika

Analisis Algoritma Naive Bayes Untuk Prediksi Kepuasan Layanan Akademik Berbasis Data Multibahasa Oktafiani, Dewi; Putra, Tommy Dwi; Kusumastuti, Rajnaparamitha
Jurnal Algoritme Vol 5 No 3 (2025): Oktober 2025 || Algoritme Jurnal Mahasiswa Teknik Informatika
Publisher : Program Studi Teknik Informatika Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/algoritme.v5i3.13456

Abstract

The quality of academic services greatly influences student satisfaction. This study predicts student satisfaction with academic services using a Naïve Bayes algorithm based on multilingual data. Data from 213 students across three departments at STMIK AMIKOM Surakarta cover five key service aspects. Student comments were processed through text preprocessing and TF-IDF weighting, then tested on both Indonesian and English-translated texts. The results showed a significant difference: the Indonesian model achieved 67.44% accuracy, 0.68 precision, 0.65 recall, and 0.66 F1-score, while the English version improved to 83.72% accuracy, 0.84 precision, 0.82 recall, and 0.83 F1-score. Statistical tests confirmed this difference as significant. The findings highlight that English NLP tools are more mature and provide empirical contributions to improving the quality of academic services in higher education.
Implementasi Groq AI untuk Otomatisasi Feedback pada Website Evaluasi Kinerja Dosen Kusumastuti, Rajnaparamitha; Oktafiani, Dewi; Dwi Putra, Tommy
Jurnal Algoritme Vol 5 No 3 (2025): Oktober 2025 || Algoritme Jurnal Mahasiswa Teknik Informatika
Publisher : Program Studi Teknik Informatika Universitas Multi Data Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35957/algoritme.v5i3.13458

Abstract

Lecturer performance evaluation is essential to maintain the quality of higher education, yet traditional methods often lack objectivity and provide limited feedback. This study designed a web-based evaluation system using the Simple Additive Weighting (SAW) method for decision-making, integrated with Groq AI to generate automatic feedback from students. The system was developed with a prototype approach using the Flask framework and tested on 10 courses with a total of 250 randomly selected respondents. Instrument reliability was confirmed using Cronbach’s Alpha (α = 0.84), indicating a high level of reliability. System speed evaluation through 40 trials showed an average processing time of 0.564 seconds. User satisfaction was measured with a 1–4 Likert scale and converted using the Percent of Maximum Possible (POMP), resulting in a 92.4% satisfaction rate. The AI feature successfully provided automated feedback without manual intervention, significantly improving efficiency and effectiveness. These results demonstrate that integrating SAW with Groq AI enhances objectivity, speed, and quality in lecturer performance evaluation.