Skripsi

KNOWLEDGE DISCOVERY BERDASARKAN ANALISIS SENTIMEN TERHADAP PERSEPSI PUBLIK TENTANG GENERATIVE AI DI X

---

Generative Artificial Intelligence (GenAI) has developed rapidly with various applications that facilitate content production but raise concerns regarding ethics, privacy, and copyright. Public perception of this technology is crucial to understand as it influences social acceptance and the direction of policy development. This study aims to map public sentiment toward GenAI on the social media platform X (Twitter) using a knowledge discovery approach that integrates topic modelling and Aspect-Based Sentiment Analysis (ABSA). A total of 111,675 tweets were collected from June 23, 2024, to June 23, 2025, using a daily crawling method. Five topic modelling algorithms were applied, namely BERTopic, Top2Vec, Latent Dirichlet Allocation (LDA), Latent Semantic Analysis (LSA), and Non-negative Matrix Factorisation (NMF). The results were evaluated using the coherence metrics C_V, UMass, UCI, and NPMI. The study's results identified 11 main topics with varying sentiment distributions. The art and ethics topics tended to trigger negative sentiment, while innovation and cloud were more dominantly positive, and the model and education topics were dominated by neutral sentiment. The integration of topic modelling and ABSA provides a deeper understanding of the main issues emerging in society. It can serve as a basis for decision-making in the development of ethical and responsible GenAI.

---

Availability

Detail Information

Series Title

-

Call Number

T1894282025

Publisher

Indralaya : Prodi Sistem Informasi, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xv, 106 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

006.380 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Informasi
Analisis Sintimen--Persepsi Publik

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment