Beneng, Juan Dastyn Pratama (2023) Analisis Sentimen Pengguna Twitter Terhadap AI-Generated Art Menggunakan Metode Naive Bayes Dan Neural Network Classifier. Undergraduate thesis, Fakultas Teknologi Informasi Universitas Merdeka Malang.

Preview	Text HALAMAN AWAL.pdf Download (1MB) | Preview
Preview	Text BAB I.pdf Download (84kB) | Preview
	Text BAB II.pdf Restricted to Repository staff only Download (431kB)
	Text BAB III.pdf Restricted to Repository staff only Download (672kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (1MB)
	Text BAB V.pdf Restricted to Repository staff only Download (65kB)
Preview	Text DAFTAR PUSTAKA.pdf Download (204kB) | Preview
Preview	Text HASIL CEK PLAGIASI.pdf Download (189kB) | Preview

Abstract

Social media platforms such as Twitter have become valuable sources of data for understanding public opinions and sentiments. This research aims to analyze sentiment in Twitter data using two popular classification algorithms: Neural Network and Naive Bayes. The research methodology involves collecting a large dataset of Twitter posts related to a specific topic of interest. The collected tweets are then processed by removing irrelevant elements such as URLs, mentions, and special characters, and employing tokenization and stemming techniques to extract essential words. The dataset is divided into training and testing sets for sentiment analysis. The Neural Network model is trained on the training set using techniques like word embedding and deep learning architectures to capture the complex relationships between words and sentiments. Meanwhile, the Naive Bayes classifier is also trained on the training set, assuming independence among features. The performance of the models is evaluated using various metrics such as accuracy, precision, recall, and f1-score. The results are compared to determine the effectiveness and efficiency of each algorithm in classifying Twitter sentiments. The findings of this research indicate that both the Neural Network and Naive Bayes models can effectively analyze sentiment in Twitter. The Neural Network model performs better in capturing complex sentiment patterns due to its ability to learn intricate representations, while the Naive Bayes classifier excels in simplicity and computational efficiency. This research has significant implications as it provides valuable insights into public sentiment on social media platforms. The results can be applied in various applications such as brand reputation management, public opinion monitoring, and market analysis. Additionally, this study contributes to the development of sentiment analysis by exploring the effectiveness of Neural Network and Naive Bayes algorithms in Twitter sentiment classification.

Actions (login required)

View Item