IMPLEMENTASI ALGORITMA NAÏVE BAYES UNTUK MENGANALISIS SENTIMEN OPINI PENGGUNA TWITTER TERHADAP KOREAN POP (KPOP)

Tiara, Reisaona (2023) IMPLEMENTASI ALGORITMA NAÏVE BAYES UNTUK MENGANALISIS SENTIMEN OPINI PENGGUNA TWITTER TERHADAP KOREAN POP (KPOP). Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (876kB)
	Text Abstrak.pdf Download (195kB)
	Text Abstract.pdf Download (192kB)
	Text BAB I.pdf Download (212kB)
	Text BAB II.pdf Download (371kB)
	Text BAB III.pdf Download (528kB)
	Text BAB IV.pdf Restricted to Registered users only Download (749kB) | Request a copy
	Text BAB V.pdf Download (194kB)
	Text DAFTAR PUSTAKA.pdf Download (335kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (599kB) | Request a copy

Abstract

Twitter is currently one of the platforms favored by many social media users. Users can exchange information, include any hashtags to make it easier to find information, and even talk about current topics. One of the most popular current topics is Kpop. The popularity of KPop in Indonesia is caused by the younger generation who think kpop has a positive impact on their lives in terms of emotional and inspiration. But there are also many people who think kpop has a negative impact. In this study, researchers took 3.000 tweet data with the keyword Kpop from twitter which will then be processed to see the sentiment of twitter users. Pro and con views on twitter will be used as a data source to monitor the results of feedback from the public on opinions about KPop. The purpose of the research is to find out the process of analyzing the sentiment of twitter users' opinions on KPop using a classification algorithm, namely Naïve Bayes and get the performance results of the Naïve Bayes Algorithm model of public sentiment towards Kpop on twitter social media that is good and appropriate. Data is processed using the Naïve Bayes classification algorithm in machine learning. The classification results are divided into 3 classes: positive, negative, and neutral. Word weighting uses TFIDF with 5 scenarios, namely the minimum number of times the word is repeated to be processed. In addition, one of the outputs of this research is the Naïve Bayes algorithm that is able to perform sentiment analysis with the highest performance results in scenario 2 of 59% accuracy, 62% precision, 51% recall, and 51% F1-Score. Keywords: Sentiment Analysis, KPop, Phyton, Naïve Bayes, Twitter

Actions (login required)

View Item