Akademik Veri Yönetim Sistemi
9th International Artificial Intelligence and Data Processing Symposium, IDAP 2025, Malatya, Türkiye, 6 - 07 Eylül 2025, (Tam Metin Bildiri)
Spoiler contents reduce the enjoyment of watching films or series. To address this issue many studies have focused on detecting spoilers in comments and reviews. However, spoiler detection is still challenging. In this work, we evaluate the effect of various machine learning approaches for spoiler detection using transformer-based embedding. Our approach consists of two main stages. First, a transformer-based model is employed to generate embeddings from reviews. Second, machine learning methods are applied to classify the embedded representations. The dataset consists of movie reviews from IMDb. The comparison of the methods is conducted using different evaluation metrics. Experimental results show that SVM and LR achieved the highest accuracy.