ANALISIS SENTIMEN PENDAPAT MASYARAKAT TERHADAP BANSER DAN FRONT PEMBELA ISLAM (FPI) BERDASARKAN OPINI DARI TWITTER MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER

Alfian Nurul Humaida, NIM.: 16650093 (2020) ANALISIS SENTIMEN PENDAPAT MASYARAKAT TERHADAP BANSER DAN FRONT PEMBELA ISLAM (FPI) BERDASARKAN OPINI DARI TWITTER MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (ANALISIS SENTIMEN PENDAPAT MASYARAKAT TERHADAP BANSER DAN FRONT PEMBELA ISLAM (FPI) BERDASARKAN OPINI DARI TWITTER MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER) 16650093_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (ANALISIS SENTIMEN PENDAPAT MASYARAKAT TERHADAP BANSER DAN FRONT PEMBELA ISLAM (FPI) BERDASARKAN OPINI DARI TWITTER MENGGUNAKAN METODE NAIVE BAYES CLASSIFIER) 16650093_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

Islam in Indonesia has its own religion, both in terms of the characteristics of da'wah, people and religious symbols. These differences of understanding can encourage the emergence of different religious groups or movements both in terms of religious aspirations, religious activities, social activities, economic and even political activities. In Indonesia there are organizations that have almost the same function but have different ideologies that are religious ideologies. Like the BANSER (Multipurpose Ansor Troops) organization part of NU, the fighting wing of the FPI (Islamic Defenders Front). This research classifies the data taken from Twitter for 10061 data to be analyzed using the Naïve Bayes Classifier method using TF-IDF weighing as well as split validation and confusion matrix for accuracy calculations. The split validation ratio used was 90:10 which then resulted in a 82.10% accuracy for the Banser and 79.96% for the FPI. The results of the implementation in the test data a total of 1878 tweet data resulted in a 22% negative sentiment classification with 209 data and positive sentiments as much as 78% with 742 of data for the Banser implementation. As for FPI data, there are 49.6% negative sentiment classification results with 460 data and positive sentiments as much as 50.4% with 467 data.

Share this knowledge with your friends :

Actions (login required)

View Item