ANALISIS SENTIMEN TERHADAP RESESI 2023 PADA MEDIA SOSIAL TWITTER MENGGUNAKAN METODE NAÏVE BAYES DENGAN SELEKSI FITUR INFORMATION GAIN

Hafidz Ashabi, . (2024) ANALISIS SENTIMEN TERHADAP RESESI 2023 PADA MEDIA SOSIAL TWITTER MENGGUNAKAN METODE NAÏVE BAYES DENGAN SELEKSI FITUR INFORMATION GAIN. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.

	Text ABSTRAK.pdf Download (235kB)
	Text AWAL.pdf Download (526kB)
	Text BAB I.pdf Restricted to Repository UPNVJ Only Download (251kB)
	Text BAB II.pdf Restricted to Repository UPNVJ Only Download (387kB)
	Text BAB III.pdf Restricted to Repository UPNVJ Only Download (344kB)
	Text BAB IV.pdf Restricted to Repository UPNVJ Only Download (4MB)
	Text BAB V.pdf Download (237kB)
	Text DAFTAR PUSTAKA.pdf Download (239kB)
	Text RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (123kB)
	Text LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (766kB)
	Text HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (513kB)
	Text ARTIKEL KI.pdf Restricted to Repository staff only Download (4MB)

Abstract

In facing global economic dynamics, recession becomes critical. Social media, especially Twitter, has become the main channel for expression and sharing views regarding economic conditions. This research aims to conduct sentiment analysis regarding the 2023 recession based on data taken from conversations on Twitter social media. The sentiment analysis method used is Naïve Bayes, which is known for its coolness in handling text data. In addition, Information Gain feature selection is applied to select the most informative words or features in predicting sentiment. Information Gain allows us to evaluate the relative contribution of each word to sentiment determination. The data used in this research consists of a large number of tweets taken in a period relevant to the context of the 2023 recession. The data processing process involves stages such as tokenization, removal of stop words, and text normalization. This research will also compare the use of Information Gain and not using Information Gain as feature selection. Data collection will be carried out by crawling using the R programming language and integrated with the API provided by Twitter. 2. The research results show that there is an improvement in the Naïve Bayes model when using the Information Gain feature selection with a top ranking value of '>0.01', namely accuracy 0.96, recall 1, precision 0.93, f1 score 0.96 and specificity 0.93 compared to before without Information gain feature selection, namely accuracy 0.92, recall 1, precision 0.85, f1 score 0.91 and specificity 1.

Item Type:	Thesis (Skripsi)
Additional Information:	[No.Panggil: 1910511113] [Pembimbing: Nur Hafifah Matondang] [Penguji 1: Ermatita] [Penguji 2: Kraugusteeliana]
Uncontrolled Keywords:	Sentiment Analysis, Twitter, Naïve Bayes, Information Gain
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software
Divisions:	Fakultas Ilmu Komputer > Program Studi Informatika (S1)
Depositing User:	Hafidz Ashabi Muhammad
Date Deposited:	25 Mar 2024 07:06
Last Modified:	25 Mar 2024 07:18
URI:	http://repository.upnvj.ac.id/id/eprint/29217

Actions (login required)

View Item