Deep learning with optimization for social media fake profile detection

Abstract The rapid increase in fake social media accounts, particularly X platform (Twitter), has been a significant issue. Fake profiles can deceive people or groups of people, disseminate falsehoods, destroy credibility, sell fake news, support evil actions, and control online interactions. Fake accounts are highly developed, thus necessitating automated, sophisticated methods to detect them manually. This paper presents a powerful deep learning model to identify false profiles in multimodal Twitter data. The architecture fuses LSTM to analyse text, CNN to analyse visual information, and ANN for metadata characteristics with Adam optimizer setting a state-of-the-art accuracy of 97.2060%, Precision: 0.9874, Recall: 0.9682, and F1-Score: 0.9777. The model does a full analysis of account activity, behavior patterns, content, and network behavior to find real and fake accounts more quickly. This paper proposes a model utilizing diverse data modalities to analyze the influence of individual characteristics and combinatorial features on detection accuracy. The proposed model improves the sphere of fake profile detection because it suggests a universal and scalable solution that will strengthen online trust and the security of platforms. It also preconditions more research and more appropriate social media rules.