Identification of Clickbait Video Titles in Indonesian Language

Clickbait on social media platform is an important problem, especially on video based content media social. Clickbait is defined as titles designed to lure users into clicking by using exaggerated titles. This phenomenon not only impacts traffic and ad revenue (AdSense) for content creators but also has negative effects such as reduced viewer trust, low-quality content, and even the spread of propaganda. This study aims to classify whether title of a video published on social media platform is a clickbait or not. We use the following classification methods: Support Vector Machine (SVM), Random Forest (RF), and Long Short-Term Memory (LSTM) were applied to our dataset. Results show that the tuned SVM achieved an F1-score of 78.13%, Random Forest demonstrated a competitive F1-score of 75.47%, and LSTM with an F1-score of 76.36%. However, each model has its limitations, such as difficulty with descriptive titles (SVM), long titles (RF), and ambiguous context (LSTM). This study contributes by providing insights into the effectiveness and challenges of implementing various models for detecting clickbait in Indonesian Video titles.