IoT ecosystem for edge-based stream analytics and complex event processing (Case Study Healthcare)

Описание: This video explains a case study that I choose to explore the connection between IoT stream analytics, historical data, complex event processing, semantic enrichment, and edge computing in the field of health data.
The motivation to extend this project to an end-to-end architecture ecosystem was based on missing guides and examples on the internet for IoT architecture implementation (everything is nowadays in the cloud). I saw that different domain experts usually realize such an example with teams consisting of specialists in semantics, CEP, big data architecture, IoT, and machine learning, as well as different programming language requirements. My motivation was to build and share this architecture as an example for interested people to dive deeper into the possibilities of realtime CEP stream processing and explore new use cases.
I am happy to present you with an approach that can be realized by one data scientist, with pre-knowledge in python, SQL, and a bit of big data architecture design.

Note on future work details:
The anomaly detection was an exploration in the technique of the HTM algorithm, where more efficient approaches can be found. Furthermore, does the existing state of the art 2D-CNN models achieve way better results at the classification level.

I hope you enjoyed the video.

I am excited for your feedback with a comment or like:)

My contact for more details or code examples: [email protected]

HTM introduction video:    • HTM Overview (Episode 0)  

Setup guide: Coming soon
Code: Coming soon

Special credits go to my supervisor Prof. Fatos Xhafa who taught and supported me to realize this project - thank you