Custom online anomaly detection for stream of data from different sources

Job ID: 34125801

Budget: $250 – $750 USD

? Heycan you help me with
design a custom online anomaly detection for stream of data from different sources? I have a proposed system architecture that explain the system in high level and I need to be implemented in unsupervised manner. The skill needed is anomaly detection ( HalfSpaceTree using RiverML) , model explanation and selection using SHAP. Also Using PSO algorithm to optimise hyperparameter for a selected anomaly detection model. If hyperparameter give better result it will replace the current model that used in online anomaly detection. If you welling to work on this and avalible I would like to talk more and explain the design and I get your feedback of what is possible and what is not. I am looking to hear from you. Thanks

https://riverml.xyz/0.11.1/
This is riverML lib

https://github.com/slundberg/shap