Analisis Sentimen pada Komen Twitter Pawang Hujan Mandalika dengan Support Vector Machine (SVM) dan Naïve Bayes

Rahmat Satria Buana (Universitas Nusa Mandiri)
Windu Gata (Universitas Nusa Mandiri)
Alda Zevana Putri Widodo (Universitas Nusa Mandiri)
Hendra Setiawan (STMIK Bani Saleh)
Khairunisa Hilyati (Universitas Nusa Mandiri)

Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi)
Vol 7 No 2 (2023): APRIL-JUNE 2023

Indonesia is a country with a majority Muslim population, but Indonesia is also rich in culture influenced by previous religions such as Hinduism and Buddhism. The rain handler by Muslims is considered shirk, but on the other hand also considers this ancestral culture that has always existed. With the advancement of human technology, it is easier to express opinions or opinions regarding a topic that is currently being discussed, for example, regarding the rain handler who acted at the Mandalika Circuit some time ago through social media. Twitter is one of the social media that is used as a forum to accommodate these opinions. In this study, the CRISP-DM (Cross Industry Standard Process for Data Mining) data mining methodology was used with Rapid Miner Version 9.10 with the Support Vector Machine (SVM) classification method and nave Bayes with SMOTE to improve accuracy. The application of the SVM method is 97.71 % with AUC 0.997 Positive Comments, and Using Naïve Bayes, the accuracy obtained is 93.41% accuracy with AUC 0.973 Positive Comments.