Penerapan Algoritma Support Vector Machine Klasifikasi Komentar Youtube berdasarkan Emosi Netizen (Teks Analisis: Weird Genius – Lathi (Ft Sara Fajira))

Alpha, Riski Pajarianto (2021) Penerapan Algoritma Support Vector Machine Klasifikasi Komentar Youtube berdasarkan Emosi Netizen (Teks Analisis: Weird Genius – Lathi (Ft Sara Fajira)). Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text COVER.pdf Download (602kB)
	Text Abstract.pdf Download (61kB)
	Text ABSTRAK.pdf Download (62kB)
	Text BAB I.pdf Download (256kB)
	Text BAB II.pdf Download (297kB)
	Text BAB III.pdf Download (198kB)
	Text BAB IV.pdf Restricted to Registered users only Download (230kB) | Request a copy
	Text BAB V.pdf Download (62kB)
	Text DAFTAR PUSTAKA.pdf Download (142kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (85kB) | Request a copy

Abstract

In this modern era, information technology is one of the things that cannot be separated from humans, many people use the internet to fill their spare time, including watching or listening to music on online videos. YouTube is currently a platform that is often used, the YouTube platform has various comments containing opinions, stories, and emotions. Therefore this study aims to explore the potential for YouTube comments and potential comments on the Weird Genius ft Sarafajira video entitled Lathi which has received an award from Google for the most wanted song in 2020 and also won 3 (three) Indonesian Music Award (AMI). in 2020. Emotion categories are divided into six emotions, namely: happiness, anger, fear, sadness, disgust, and surprise. for emotional classification needs, comments are obtained from crawling using the YouTube comment downloader, after comments are obtained, the preprocessing stage is carried out and Comment labeling was carried out using the NRC-Emotion Lexicon word list and emotion classifying using the Support Vector Machine (SVM) algorithm to determine its accuracy. Comments obtained were 31,068 YouTube comments and preprocessing and labeling were carried out which resulted in 8113 data containing emotions. The results of this study indicate that the most dominant emotion is Happy as many as 4992 comments and by using SVM can be an accuracy of 76% with a ratio of 80% and 20%. Keywords: emotional classification, NRC-Emotion Lexicon, Support Vector Machine, YouTube

Actions (login required)

View Item