I received a lovely surprise package this month. An invitation to the University of Manchester to give a seminar to the Decision and Cognitive Sciences (DCS) research centre. What amazes me is how this opportunity came in the most unimaginable way. It reinforces to me the power of networking and taking daring steps along the lines of one’s dreams. I will save the story for a later time.
The seminar will hold on the 25th of September 2019 from 12:30 to 14:00. My talk is titled: “Intelligent data stream processing in distributed systems: the case of anomaly detection in images”. It will be an hour long seminar – 50 minutes talk and 10 minutes for questions.
Here is the abstract of my talk.
Internet of Things (IoT) ubiquitous sensors and devices are generating massive data streams continuously. These streams need to be processed on-the-fly to extract knowledge for several applications like video surveillance, autonomous vehicles, smart city, web monitoring, etc. The existing approach for data stream processing is designed for centralised systems where all the data is sent to the data centres for storage and analytics. However, it is often not feasible to migrate all the data to the cloud for cost, performance and privacy concerns. In decentralised systems like IoT networks, other agents like end devices, edge nodes, and cloudlets can cooperatively participate in the processing pipeline. To achieve this, the existing frameworks and techniques for centralised systems would need to be completely redesigned or modified to meet the demands of decentralised environments. In this talk, we discuss the unique challenges of intelligent data stream processing in decentralised systems. Emerging research on techniques and frameworks for intelligent stream processing in decentralised networks will be explored. We present a motivating use case of data stream processing in a decentralised system with anomaly detection in image streams.
I am earnestly looking forward to it.
Great work. I am also working on adversarial machine learning. It has to do with anomaly detection for training models. However, U am not using stream data processing techniques.
Hi Peter,
That’s great. It’ll be nice to know more about how you’re tackling anomaly detection with adversarial machine learning techniques.
You finally pulled this off. Congrats Maryleen! The long hours of class in that professor’s lectures have paid off greatly.
Your blogs are insightful and helpful. Keep it up and best of luck in your future endeavours.
Hi Nii,
Yes, indeed it did pay off. Thanks for visiting my blog and for the nice comments.