Real-Time Outlier Detection in Streaming Data : A Deep Learning-Based Approach

Outliers are observations that deviate significantly from the norm, and their detection has been a critical topic in various research areas and application domains, such as video surveillance, network intrusion detection, and disease outbreak detection. In recent years, deep learning-based techniques have shown to outperform machine learning and shallow approaches for outlier detection in streaming data, which are large and complex datasets. However, developing an effective and appropriate model for outlier detection is challenging due to the dynamic nature and variations of real-world applications and data.In this research, we propose a novel deep neural network (DNN) model for outlier detection in streaming data. Our model is developed with multiple hidden layers to improve feature abstraction and capabilities. We evaluate our proposed model on four real-world outlier benchmark datasets available at the UCI repository and compare its performance with state-of-the-art approaches. Our experiments demonstrate that our proposed model outperforms both machine learning algorithms and deep learning competitors, resulting in significant performance gains. In particular, our proposed approach achieves much higher accuracy, recall, and f1-score rates of 99.63%, 99.014%, and 99.437%, respectively, compared to other algorithms.Our proposed deep learning-based approach provides an effective and efficient solution for real-time outlier detection in streaming data. The experimental results show its superior performance compared to existing methods, which makes it a valuable contribution to the field of outlier detection