USI - Email

Archive / INF Seminars / INF_2022_09_29_Hristijan_Gjoreski

USI - Email

	Ambient Intelligence and Machine Learning

Host: Prof. Marc Langheinrich

Thursday

29.09

USI Campus EST, room D1.14, Sector D

13:30-14:30


	Hristijan Gjoreski Ss. Cyril and Methodius University in Skopje, North Macedonia

Biography:

Hristijan Gjoreski is Associate Professor at the Ss. Cyril and Methodius University in Skopje, North Macedonia, and is lead data scientist at Emteq Labs, UK. He finished his PhD in 2015 at the Jozef Stefan Institute in Slovenia, and was postdoctoral researcher in 2017 at University of Sussex, UK. His scientific and research experience is in the domains of Machine Learning and applied Artificial Intelligence. He has specialized in development of machine-learning algorithms in the areas of wearable computing, e-health, activity recognition and affective computing. He has received the award “Best Young Scientist” for 2016 from the President of Republic of Macedonia. Hristijan and his teams have won 3 international machine learning competitions for human activity recognition with wearable sensors. He has participated in more than 12 international projects, and he is currently the coordinator of the WideHealth European H2020 project.

Abstract:
The recent technological advancements in sensor miniaturization allowed numerous applications in the areas of ambient intelligence, ubiquitous computing, mobile health etc. This talk will focus on few important machine-learning applications in with wearable devices such as smartphones, smartwatches, smart wristbands, etc. The first part of the talk will introduce the whole process of development of a machine learning pipeline for human activity recognition with wearable accelerometers. Numerous technical steps in the creation of the classification model will be discussed, such as: sensor data filtering, data segmentation, feature engineering, training a classification model and its evaluation. Additionally, some ideas and recent trends will be presented, including deep learning approaches, transfer learning, and unsupervised learning. The second part of the talk will focus on other ML applications in wearable computing such as: calorie expenditure estimation, fall detection, stress and arousal recognition. Each of these applications will be individually discussed and appropriate ML solutions will be presented.