Publications by authors named "Muhammad Ali Chattha"
Article Synopsis
- Unsupervised anomaly detection is crucial for managing streaming data from smart devices, helping to prevent machine downtime through real-time monitoring of various sensor data.
- Different types of data may call for different anomaly detection methods, with some benefiting from statistical techniques and others from deep learning approaches.
- The paper introduces FuseAD, a novel technique that merges statistical and deep learning methods, demonstrating improved performance on benchmark datasets compared to existing methods.
