Analisis Performa Algoritma NBC, DT, SVM dalam Klasifikasi Data Ulasan Pengunjung Candi Borobudur Berbasis CRISP-DM

Support Vector Machine (SVM) emerges as the optimal algorithm for analyzing visitor sentiment at Borobudur Temple, outperforming Naïve Bayes Classifier (NBC) and Decision Tree (DT). SVM achieved 99.41% accuracy, 100% precision, and a perfect Area Under Curve (AUC) score of 1.00, significantly surpassing NBC's 95.6% accuracy and DT's 94.4%. The study, based on 3850 Tripadvisor reviews, applied the CRISP-DM framework to evaluate algorithm performance using metrics like confusion matrix values, AUC, and ROC. SVM's dominance was further validated by a T-test value of 0.994, indicating its superior effectiveness in sentiment analysis for this UNESCO World Heritage Site.

Singgalen, Y. A. (2022). Analisis Performa Algoritma NBC, DT, SVM dalam Klasifikasi Data Ulasan Pengunjung Candi Borobudur Berbasis CRISP-DM. Building of Informatics, Technology and Science (BITS), 4(3). https://doi.org/10.47065/bits.v4i3.2766