WePaMaDM-Outlier Detection: Weighted Outlier Detection using Pattern Approaches for Mass Data Mining
This work addresses outlier detection for applications like surveillance and fault detection, but it appears incremental as it adapts existing techniques to specific domains.
The paper tackles the challenge of detecting outliers in mass data mining by proposing the WePaMaDM-Outlier Detection method, which is domain-dependent and adapts solutions for specific problem formulations, but does not report concrete numerical results.
Weighted Outlier Detection is a method for identifying unusual or anomalous data points in a dataset, which can be caused by various factors like human error, fraud, or equipment malfunctions. Detecting outliers can reveal vital information about system faults, fraudulent activities, and patterns in the data, assisting experts in addressing the root causes of these anomalies. However,creating a model of normal data patterns to identify outliers can be challenging due to the nature of input data, labeled data availability, and specific requirements of the problem. This article proposed the WePaMaDM-Outlier Detection with distinct mass data mining domain, demonstrating that such techniques are domain-dependent and usually developed for specific problem formulations. Nevertheless, similar domains can adapt solutions with modifications. This work also investigates the significance of data modeling in outlier detection techniques in surveillance, fault detection, and trend analysis, also referred to as novelty detection, a semisupervised task where the algorithm learns to recognize abnormality while being taught the normal class.