Intelligent Framework for Detecting Predatory Publishing Venues

Predatory publishing venues publish questionable articles and pose a global threat to the integrity quality of scientific literature. They have given rise dark side scholarly their effects reached political, societal, economic, health aspects. Given consequences proliferation, several solutions been developed help detect them; however, these are manual time-consuming. While researchers, students, readers in need tool that automatically detects predatory violations, this study, we proposed an intelligent framework can violations using different artificial intelligence techniques. This work contributes through following: (1) creating dataset 9,866 journals annotated as legitimate, (2) proposing for classifying venue legitimate or predatory, with appropriate reasoning. Our was evaluated seven machine learning deep models, including Support Vector Machine (SVM), K-Nearest Neighbors (KNN), Neural Networks (NNs), Long short-term memory (LSTM), Convolutional Network (CNN), Bidirectional Encoders from Transformers (BERT), A Lite BERT (ALBERT), feature representation The results showed CNN model outperformed other models journal classification task, F1 score 0.96. For reasoning provisioning SVM achieved best micro 0.67.