"Anomaly Detection Benchmark Datasets (ODDS Subset) used in D2H-AD"
Abstract
"This repository contains the pre-processed benchmark datasets used to evaluate the D2H-AD model presented in the associated IEEE Access manuscript. The datasets include:WBC: Wisconsin Breast Cancer (Diagnostic)MNIST: Handwritten digits (subset for anomaly detection)CARDIO: Cardiotocography measurementsLYMPHO: Lymphography domain dataSATI2: Satimage-2 (Satellite imagery)Source: These datasets were originally sourced from the ODDS (Outlier Detection DataSets) Library [Rayana, 2016]. They are provided here in the specific .csv formats used in the study to ensure experimental reproducibility, as the original repository links may be intermittent or inaccessible.Citations: Rayana, S. \"ODDS Library [http:\/\/odds. cs. stonybrook. edu].\" Stony Brook University, Department of Computer Science, Stony Brook, NY (2016)."
Fields of Study & Subjects
Anomaly Detection Techniques and Applications
Adversarial Robustness in Machine Learning
Artificial Immune Systems Applications
Publication details
- Journal / source
- IEEE DataPort
- Publication date
- 19 January 2026
- Document type
- Dataset