Python code for abnormal detection using Support Vector Data Description (SVDD)

One-class classifiers for anomaly detection (outlier detection)

2D residual U-Net (ResUNet) and a lead combiner (LC) for 12-lead ECG Abnormality Classification

Built an evaluation model based on user authentication data to predict the risk of current user authentication behavior.

Identified risk accounts involved in transaction fraud based on account data.

NetSniff-Guard is an advanced network security tool that captures live network packets and detects anomalies using machine learning techniques. It analyzes traffic in real-time based on multiple features including packet size, protocol, timing patterns, and flow behavior. The system provides immediate alerts for suspicious activities.

PyTorch unofficial implements `PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization` paper.

实现高质量、可商业化落地的预测性维护（Predictive Maintenance, PdM）系统，通过对拖拉机运行数据进行实时监控和深度分析，实现对关键部件故障的提前预测、健康状态的精准评估和智能化维保决策，从而最大化设备可用性、降低非计划停机时间，并优化维护资源配置。

Code Reproduction of the essay Isolation Forest

The ECG Detection with Deep Learning project employs Convolutional Neural Networks to automatically analyze Electrocardiogram (ECG) data, facilitating precise detection of cardiac abnormalities and enhancing diagnostic accuracy.