zwang-datascience / MVAD_Bayesian

Python implementation of the proposed multi-view anomaly detector
Apache License 2.0
5 stars 2 forks source link

Inductive Multi-View Anomaly Detection

Model implementation for IJCAI 2020 paper "Towards a hierachical Bayesian model of multi-view anomaly detection" by Zhen Wang and Chao Lan from University of Wyoming.

Different from all exsiting multi-view anomaly detection techniques, this work is the first attempt to detect multi-view outliers/anomlies under semi-supervised scenario via a Bayesian model of inductive learning. For details, refer to our paper.

Dataset Preparation

Download the below raw datasets, then preprocess them(, mainly feature rescaling) and generate the mutli-view outliers following the strategy mentioned in our paper or related works.

Citation

If this code is useful for your research, please cite our paper:

Wang, Zhen, and Chao Lan. "Towards a Hierarchical Bayesian Model of Multi-View Anomaly Detection." In IJCAI, pp. 2420-2426. 2020.