ANALISIS SENTIMEN TWITTER TERHADAP PROFESI TUKANG PARKIR MENGGUNAKAN METODE NAÏVE BAYES

Muhammad Sholihuddin Nur, NIM.: 16650044 (2020) ANALISIS SENTIMEN TWITTER TERHADAP PROFESI TUKANG PARKIR MENGGUNAKAN METODE NAÏVE BAYES. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (ANALISIS SENTIMEN TWITTER TERHADAP PROFESI TUKANG PARKIR MENGGUNAKAN METODE NAÏVE BAYES) 16650044_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (ANALISIS SENTIMEN TWITTER TERHADAP PROFESI TUKANG PARKIR MENGGUNAKAN METODE NAÏVE BAYES) 16650044_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

The existence of the parking attendant profession is a topic that can be quite energy-consuming when discussed, both in terms of functionality, value and price of services provided, and also the urgency of its existence, because there is no research that discusses it. Therefore, a study of sentiment analysts on the parking attendant profession is carried out. This study classifies 1357 data taken from Twitter to be analyzed by the Naïve Bayes method with TF-IDF as a term weighting model, and uses a split validation and confusion matrix as the evaluation stage of the model, then the next implementation of 1000 new data to predict. The split validation ratio used in each model shows different accuracy using the evaluation of the confusion matrix. Model with a split ratio of 90:10 producing an accuracy of 82.35%, a split ratio of 80:20 producing an accuracy of 79.04%, while a split ratio of 70:30 resulting in an accuracy of 78.92%. The results of the implementation to the test data of 1000 tweets taken in the span of July 7-12 using the model produces positif tweet predictions of 28.9% and negatif tweet predictions of 71.1%.

Share this knowledge with your friends :

Actions (login required)

View Item