Advanced Health Misinformation Detection Through Hybrid CNN-LSTM Models Informed by the Elaboration Likelihood Model (ELM)
Mkululi Sikosana, Sean Maudsley-Barton, Oluwaseun Ajao · 2025
Health misinformation during the COVID-19 pandemic has significantly challenged public health efforts globally. The proposed addition of the Elaboration Likelihood Model (ELM) to an existing hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model enhances health misinformation detection on social media. The model enhances the detection accuracy and reliability of misinformation classification by integrating ELM-based features such as text readability, sentiment polarity, and heuristic cues (e.g., punctuation frequency). The enhanced model (text + ELM features) achieved an accuracy of 97.37%, precision of 96.88%, recall of 98.50%, F1-score of 97.41%, and ROC-AUC of 99.50%. A combined model (text + ELM + additional features) further improved performance, achieving a precision of 98.88%, recall of 99.80%, F1-score of 99.41%, and ROC-AUC of 99.80%. These findings highlight the value of ELM features in improving detection performance, offering valuable contextual information. To the best of our knowledge, this is the first empirical study to embed structured ELM-derived psychological cues, both central and peripheral, into a hybrid CNN-LSTM framework for automated misinformation detection. This integration moves beyond traditional theoretical applications of ELM and demonstrates its computational relevance for social media classification tasks.