Analisis Sentimen Pengguna Twitter Terhadap Program Ibukota Nusantara Menggunakan Metode Naïve Bayes

Pratamsyah, Adam Arditya Abna and , Widi Widayat, S.Kom., M.Eng. (2025) Analisis Sentimen Pengguna Twitter Terhadap Program Ibukota Nusantara Menggunakan Metode Naïve Bayes. Skripsi thesis, Universitas Muhammadiyah Surakarta.

	PDF (Naskah Publikasi) Revisi ADAM TTD fixed.pdf Download (516kB)
	PDF (Surat Pernyataan Publikasi) UNGGAH Abna.pdf Restricted to Repository staff only Download (305kB) | Request a copy

Abstract

Twitter, as a social media platform, allows users to engage in discussions on a wide range of topics, one of which is the IKN (Ibu Kota Negara) program, a current and widely debated issue. This study aims to analyze public sentiment towards the IKN program by categorizing sentiments into two main types: positive and negative. The research employs the Naïve Bayes classification method, known for its simplicity and efficiency in text data processing. The process involves several pre-processing steps, including data cleaning, case conversion to lowercase, word normalization, stopword removal, tokenization, stemming, and data labeling for sentiment categorization. The dataset comprises 1,012 tweets, divided into 70% for training and 30% for testing. The evaluation of the model uses a Confusion Matrix, yielding an accuracy of 69%, indicating the model correctly predicts sentiment in 69% of the test data. A precision of 70% demonstrates that 70% of the tweets classified as positive or negative are accurate predictions. The recall of 69% indicates the proportion of correctly detected positive or negative tweets, while the F1-score of 60% reflects a balance between precision and recall, showcasing the model’s consistency in sentiment classification. The results indicate that the Naïve Bayes model is efficient for sentiment analysis on tweets regarding the IKN program. However, the accuracy and F1-score suggest that further improvement is possible, such as enhancing data quality or exploring alternative classification methods for better outcomes.

Item Type:	Thesis (Skripsi)
Uncontrolled Keywords:	Sentiment Analysis, Twitter, Naïve Bayes
Subjects:	H Social Sciences > HE Communications > Media Massa T Technology > T Technology (General) T Technology > TZ Technical Information
Divisions:	Fakultas Komunikasi dan Informatika > S1 Teknik Informatika
Depositing User:	ADAM ARDITYA ABNA PRATAMSYAH
Date Deposited:	18 Feb 2025 03:26
Last Modified:	18 Feb 2025 03:26
URI:	http://eprints.ums.ac.id/id/eprint/132614

Actions (login required)

View Item