An Investigation Of The ML/DL Hybrid Model For Social Media Attack Prediction And Detection

Abstract

Deep learning-based intrusion detection technology has been extensively researched in both academia & business. In order to classify challenges involving hybrid model to predict and detect attacks in social media, deep learning uses hybrid models to extract probable features from high-dimensional data. The 1D-CNN is a novel form of the CNN that can distinguish between normal and attack input in order to identify attacks Separate DL models based on 1D-CNN were created and applied to both the combined datasets and the other sub-datasets. The suggested IDS uses a deep learning model that blends CNN & LSTM algorithms, providing a hybrid approach. CNN can identify unknown threats since it is built to recognize objects.