Ubicomp'24 Workshop on Advancing Physiological Methods in HII

📢 Join us for an engaging, interactive workshop to explore the fascinating world of user experience in HII!

About This Workshop

With the advancement of pervasive technology, Human-Information Interaction (HII) has become increasingly ubiquitous. In these diverse information access devices and interfaces, it is crucial to understand and improve the user experience during human-information interaction. In recent years, we have seen a rapid uptake of physiological sensors used to estimate the cognitive aspect of the interaction. However, several challenges remain from a ubiquitous computing perspective, such as the definitions discrepancy of cognitive activities (for example, cognitive bias or information need) and the lack of standard practice for collecting and processing physiological data in information interaction.

This workshop aims to discuss and establish a common understanding of user experience in HII, discuss best practices to quantify with physiological sensing methods, and also reflect on potential applications and ethical issues.

Our workshop is also open for people who are new to the field – if you don’t have prior experience but are interested to know, our workshop is a good opportunity to start (with hands-on tutorials).

📝Register via Ubicomp website

Workshop Themes

We welcome researchers and students from diverse fields who are interested in this area to engage in in-depth discussions of the following research themes:

📍 Exploring and Defining Cognitive Activities in HII. For example:
What cognitive activities impact the interaction between humans and information, e.g., relevance, satisfaction, cognitive bias?
This includes but is not limited to web search, conversational search, social media interactions, or interactions with Large Language Models.

📍 Methods to Quantify Cognitive Activities in HII. For example:
What tools and modalities can quantify cognitive activities in HII?
What are the ground truths, and do we need them?
How can we ensure that the collected data are ecologically valid?
What are the considerations for using physiological sensors in HII settings?

📍 Application Scenarios and Impacts. For example:
What kind of applications would cognitive activity quantification enable and benefit users of information systems?
What are ethical, legal, and privacy considerations arising from employing physiological sensors in HII?

📍 Case Studies:
Realistic cases where the utilization of physiological signals has been adopted into research related to HII.

Keynote

Dr. Mahsa Salehi, Monash University

Title: Advanced methods for EEG Representation Learning and EEG Classification

Abstract: In this talk, I will present our latest methods for extracting rich information from EEG (electroencephalogram) data, one of the key physiological modalities for quantifying cognitive activity. A major challenge in EEG analysis is dealing with noise, and I will introduce a novel approach that effectively learns informative representations while maintaining robustness to noise. These robust representations can be applied to various downstream tasks, including the analysis of cognitive activities in the human brain. The second part of the talk will focus on another innovative method for classifying EEG signals in a specific application—measuring affective and motivational states as conditions for cognitive and metacognitive processes in self-regulated learning. Together, these advancements demonstrate the potential of EEG-based techniques for providing deeper insights into the human brain.

Bio: Mahsa Salehi is a Senior Lecturer and Deputy Director of Engagement in the Department of Data Science and AI at Monash University, Australia. She holds a PhD in Computer Science from the University of Melbourne and was previously a postdoctoral researcher at IBM Research Australia. Her research spans data mining and machine learning, with a focus on multi-dimensional time series analysis, anomaly detection, non-stationary distributions, and brain-inspired machine learning. Mahsa serves as an associate editor for Transactions on Knowledge Discovery from Data and has received multiple awards, including the ICDM 2022 Best Paper Runner-Up and the SIG KDD 2024 Appreciation Award. In 2016, she was recognized among the "Top 200 Most Qualified Young Researchers in Computer Science and Mathematics" by the Heidelberg Laureate Forum Foundation.

Workshop Program

13:30-14:30	Welcome and Keynote
14:30-15:30	Lighting talks and hands-on tutorials
15:30-16:00	Coffee/tea Break
16:00-16:45	Concluding Remarks