PERBANDINGAN ALGORITME K-NEAREST NEIGHBOR DAN NAÏVE BAYES PADA SENTIMENT ANALISIS DALAM KOMENTAR MEDIA SOSIAL INSTAGRAM #2019GANTIPRESIDEN

WAHYU, ROKHMANA (2019) PERBANDINGAN ALGORITME K-NEAREST NEIGHBOR DAN NAÏVE BAYES PADA SENTIMENT ANALISIS DALAM KOMENTAR MEDIA SOSIAL INSTAGRAM #2019GANTIPRESIDEN. Undergraduate Thesis thesis, Institut Telkom Purwokerto.

	Text COVER.pdf - Accepted Version Download (541kB)
Preview	Text ABSTRACT.pdf - Accepted Version Download (3kB) | Preview
Preview	Text ABSTRAK.pdf - Accepted Version Download (69kB) | Preview
Preview	Text BAB I.pdf - Accepted Version Download (285kB) | Preview
Preview	Text BAB II.pdf - Accepted Version Download (147kB) | Preview
Preview	Text BAB III.pdf - Accepted Version Download (283kB) | Preview
	Text BAB IV.pdf - Accepted Version Restricted to Registered users only Download (216kB)
Preview	Text BAB V.pdf - Accepted Version Download (4kB) | Preview
Preview	Text DAFTAR PUSTAKA.pdf - Accepted Version Download (76kB) | Preview

Abstract

ABSTRACT Instagram has been widely used by the community as one of the entertainment and information media so that many people who make it convey aspirations. The substitution of a president and vice president in 2019 serves as a means of community aspirations in # 2019GantiPresiden through comments on Instagram. The data used in this study are comments about # 2019GantiPresiden with positive and negative labels with a total of 450 data with 404 as training data and 45 as test data. This study uses the Naïve Bayes algorithm and K-Nearest Neighbor, the Naïve Bayes algorithm has the concept of probability and statistics while the K-Nearest Neighbor has the concept of the number of closest neighbors. The two algorithms are compared to determine the accuracy value of both. Obtained an accuracy value on Naïve Bayes by 73% then increased to 86% after validation at fold = 7, while K-Nearest Neighbor the highest accuracy value was 62% at k = 1 and increased to 70% after validation at fold = 7. Naïve Bayes algorithm has better performance for classification, while the K-Nearest Neighbor algorithm classification results are based on the number of k closest. Keywords - Classification, Instagram, K-Nearest Neighbor, Naïve Bayes

Actions (login required)

View Item