81 datasets found

Random Walks-based Anomaly Detection (RWAD) is commonly used to identify anomalous patterns in various applications. An intriguing characteristic of RWAD is that the input graph...

eBay product health metrics for different domain teams to monitor

The IoT testbed dataset contains data from various IoT devices under different attack conditions.

Robust anomaly detection for multivariate time series through stochastic recurrent neural network.

Image reconstruction-based anomaly detection models are widely explored in industrial visual inspection. However, existing models usually suffer from the trade-off between...

Remote Patient Monitoring (RPM) dataset, containing data from smart health devices and home sensors.

Geospatial datasets are diverse, naturally spatiotemporal, and inherently multimodal (composed of two or more distinct signal types or modalities) e.g., satellite/aerial imagery...

Anomaly detection in crowded scenes

The dataset is a collection of user messages, commands, time, and user information from the Twitch Plays Pokemon game.

This dataset is used for time-series anomaly detection. It contains multivariate time-series data from various sources, including industrial machines and sensors.

The KolektorSDD2 dataset is a dataset containing 3,335 color images of defective production items, with defects localized using pixel-wise masks.

The N-BaIoT dataset is a large dataset of network traffic traces from various IoT devices, including botnets.

The Spot-the-difference self-supervised pre-training for anomaly detection and segmentation dataset.

The CICIDS2017 dataset was collected 2 years after the UNSW NB-15 data. The collection took place for a week. On each day attacks such as: Brute Force FTP, Brute Force SSH, DoS,...

The SARAS dataset is a dataset for anomaly detection in robotic-assisted surgery. It contains 18 videos of radical prostatectomy procedures performed on artificial anatomies.

A dataset of synthetic data generated for gravitational wave detection, consisting of 400,000 noise samples and 64,000 signal samples.

Sewer-ML is a multi-label sewer defect classification dataset and benchmark.

Towards total recall in industrial anomaly detection.

The UEA archive is a collection of multivariate time series datasets where classification is of importance.

The UCR archive is a collection of univariate time series datasets where classification is of importance.