SENTIMENT ANALYSIS USING NAIVE BAYES CLASSIFIER ON ARTIFICIAL INTELLIGENCE TECHNOLOGY DEVELOPMENT

CAHYA, LINTANG DWI (2025) SENTIMENT ANALYSIS USING NAIVE BAYES CLASSIFIER ON ARTIFICIAL INTELLIGENCE TECHNOLOGY DEVELOPMENT. Tugas Akhir thesis, Informatics.

Text 5210411351_Lintang Dwi Cahya_Abstrak.pdf Download (11kB)

Abstract

The rapid advancement of Artificial Intelligence (AI) has sparked widespread discussion across various platforms, especially on social media. Public opinion on AI development is diverse, ranging from strong support to deep concern over its implications. This study aims to develop a sentiment analysis system using the Naive Bayes Classifier algorithm to classify public opinions extracted from Twitter. The research involves several stages, including data collection, text preprocessing (cleaning, tokenizing, stemming, and stopword removal), translation into English, sentiment labelling using TextBlob, and feature weighting with TF-IDF. To enhance model performance, techniques such as SMOTE for data balancing, Chi-Square for feature selection, and class prior adjustment were applied. Evaluation results show that the combination of TF-IDF, Chi-Square, class prior adjustment, and Naive Bayes achieved the highest accuracy of 73.33% with a 90:10 train-test data split. The system is also equipped with an interactive interface for text input, sentiment prediction, and data visualization. This study demonstrates that the Naive Bayes method is effective for classifying large-scale public sentiment toward AI in an efficient manner. Keywords: Sentiment Analysis, Naive Bayes Classifier, TF-IDF, Chi-Square, TextBlob.

Item Type:	Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir)
Subjects:	T Technology > T Technology (General)
Divisions:	Fakultas Sains Dan Teknologi > S1 Informatika
Depositing User:	Kaprodi S1 Informatika UTY
Date Deposited:	16 Jul 2025 02:42
Last Modified:	16 Jul 2025 02:42
URI:	http://eprints.uty.ac.id/id/eprint/18160

Actions (login required)

View Item