Day 1: Monday 5th of October
9am - 11:30am (CEST)
Biswas, S & Chowdhury, C.
Smartphone and Sensor based Health Monitoring Enabled with IoT and ML
Due to the emergence of IoT, the healthcare domain is currently undergoing a major paradigm shift toward remote and proactive healthcare. The recent pandemic caused by COVID-19 further emphasizes the need for remote and proactive health monitoring supporting the notion of social distancing. Health vitals monitoring, posture, and fall detection using widely available and affordable sensors in the form of smartphones, wearable devices, WBAN, etc. pave the way to this new paradigm of healthcare monitoring. These applications are mostly data-centric and thus, machine learning techniques are found to play a key role in data analysis and knowledge extraction. So, the tutorial will cover a 360-degree view of the problem right from the networking issues, data collection, and preparation till identification of meaningful patterns along with detailed experimental designs. The tutorial will conclude with the key insights learned and a note on the open research issues in this area.
1pm - 3pm (CEST)
Most participants of MobileHCI will have had to deal with statistics: sometimes formal statistics with tests and probabilities, sometimes informally `eyeballing data’. Even if you do not use statistics in your own work, you are likely to need to interpret the statistics in other people’s papers and reports. This tutorial is intended to help you navigate the often confusing discussions about the use and misuse of statistics, to help you gain a ‘gut feeling’ for statistics and also know when not to trust those instincts. The tutorial builds on the instructor’s recent book “Statistics for HCI” and also extensive freely available videos and other online material.
3:30pm - 6pm (CEST)
Wintersberger, P. & Holthausen, B.
Integrating Trust Measurements into Experimental Designs
Trust in automation has shown being an important construct guiding humans’ interactions with automation in a great variety of application areas with different preconditions. Potential scenarios include interaction with robots, automated vehicles, health care or manufacturing systems/devices, or predictions of artificial intelligence. The proposed tutorial provides participants a theoretical background including trust definitions, models, and measurements, and discusses how direct (standardized scales, interviews, UX curves, laddering) and indirect (behavioral, physiological) measurements can be integrated into experimental designs according to defined research goals. After the tutorial, participants will be able to integrate relevant measurements for trust in their own experimental designs to systematically assess the impact of trust-related interventions.