CBRW (Coupled Biased Random Walk) and SDRW (Subgraph Density-augmented Random Walk) are anomaly detection methods for non-IID categorical data. This repository implements these two methods in Python :)
The dataset used in the jupyter notebook can be found in https://sites.google.com/site/gspangsite/sourcecode/categoricaldata
Reference:
Pang, G., Cao, L., & Chen, L. (2021). Homophily outlier detection in non-IID categorical data. Data Mining and Knowledge Discovery, 1-62.