Abstract:
Sensor-based human activity recognition has long been a central research challenge in pervasive computing. Despite significant advancements in sensing technologies and AI methods for processing sensor data, recognising complex activities remains difficult—especially in contexts where labeled data is scarce and video-based sensing should be avoided.
Addressing this issue within smart homes holds particular promise for digital health and telemedicine. The recognition of complex activities and changes in the ability to perform them will enable the identification of digital biomarkers for the early diagnosis and continuous monitoring of medical conditions, including cognitive decline.
In this talk, I will present the core research challenges and recent progress in addressing them, with a focus on neuro-symbolic approaches and efforts to provide explainable activity predictions. In addition to the technical advancements, I will share insights from an ongoing collaboration with neurologists, highlighting the practical challenges we encounter in deploying these systems in the homes of elderly individuals.
Bio:
Claudio Bettini is professor of Distributed and Pervasive Systems in the Computer Science Department at Università degli Studi di Milano (UniMi). He leads the EveryWare Lab, a research laboratory investigating mobile and pervasive computing data management with a focus on eHealth and assistive technology applications. He was a co-founder and CSO of the Everyware Technologies spin-off and he has been for more than a decade, an affiliate research professor at the Center for Secure Information Systems at George Mason University, VA where he acted as PI of NSF funded projects related to privacy in mobile and pervasive computing. At UniMi he is currently leading efforts in projects funded by the Next generation EU initiative. He is a steering committee member of the IEEE PerCom conference, and he has been associate editor of several journals including ACM IMWUT, the IEEE Transactions on Knowledge and Data Engineering, The VLDB Journal, and the Pervasive and Mobile Computing Journal.
Link to join virtually: https://cam-ac-uk.zoom.us/j/81322468305
This talk is being recorded. If you do not wish to be seen in the recording, please avoid sitting in the front three rows of seats in the lecture theatre. Any questions asked will also be included in the recording. The recording will be made available on the Department’s webpage
