Automatic Fake News Detection in Social Networks

Bahar D., DİRİ B.

2023 Innovations in Intelligent Systems and Applications Conference, ASYU 2023, Sivas, Türkiye, 11 - 13 Ekim 2023 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/asyu58738.2023.10296553
  • Basıldığı Şehir: Sivas
  • Basıldığı Ülke: Türkiye
  • Anahtar Kelimeler: Classification Algorithms, Fake News Detection, Natural Language Processing, Text Classification
  • Yıldız Teknik Üniversitesi Adresli: Evet

Özet

With the development of technology, the spread of fake news on social networks is increasing. Many researchers and organizations have taken action to detect fake news manually or automatically. In this study, various Machine Learning Algorithms and Transformer based approaches are used to select the best performing model that can distinguish news as fake or real. In order to contribute to the Turkish literature in the field of Natural Language Processing (NLP), the dataset is specifically prepared in Turkish. The words were vectorized using Word2Vec, BERT and SBERT and classified using Machine Learning Algorithms such as Support Vector Machines, Naive Bayes, Logistic Regression, KNN and BERT/SBERT deep learning models. The highest F1 score of 0.99 was obtained from the transformer-based BERT and SBERT.