Silvia Santini |

A foundation model for electrodermal activity data

Thu, 01 Jan 2026 00:00:00 +0000

Diagnosing Constraint Awareness for Trustworthy Reinforcement Learning in MEC

Thu, 01 Jan 2026 00:00:00 +0000

Interpretability in Deep Time Series Models Demands Semantic Alignment

Thu, 01 Jan 2026 00:00:00 +0000

Keynote - Beyond Accuracy: The Good, the Bad, and the Unknown in Sensor-Based Human Behavior Modeling

Thu, 01 Jan 2026 00:00:00 +0000

Measured Enough: How Academic Knowledge Workers Negotiate Data Sharing and Control

Thu, 01 Jan 2026 00:00:00 +0000

PhD position: XAI-FinCrime

Mon, 20 Jan 2025 00:00:00 +0000

The People-Centered Computing Lab is looking for a PhD student in machine learning and explainable AI applied to financial crime detection.

Source: .

Postdoc position: XAI-FinCrime

Mon, 20 Jan 2025 00:00:00 +0000

The People-Centered Computing Lab is looking for a postdoctoral researcher in machine learning and explainable AI for financial crime detection and prevention.

Source: .

Blue Mondays and Happy Meals: Investigating the Impact of Contextual Factors on Mood States

Wed, 01 Jan 2025 00:00:00 +0000

Causally Reliable Concept Bottleneck Models

Wed, 01 Jan 2025 00:00:00 +0000

Causally Reliable Concept Bottleneck Models

Wed, 01 Jan 2025 00:00:00 +0000

Exploring Generalist Foundation Models for Time Series of Electrodermal Activity Data

Wed, 01 Jan 2025 00:00:00 +0000

Multi-Frequency Federated Learning for Human Activity Recognition Using Head-Worn Sensors

Wed, 01 Jan 2025 00:00:00 +0000

Robust PPG Authentication: The Role of Temporal, Individual, and State Variability

Wed, 01 Jan 2025 00:00:00 +0000

Unobtrusive Perceived Sleep Quality Monitoring in the Wild

Wed, 01 Jan 2025 00:00:00 +0000

Lateralization Effects in Electrodermal Activity Data Collected Using Wearable Devices

Mon, 01 Jan 2024 00:00:00 +0000

Modeling Perceived Sleep Quality Through Objective and Subjective Data

Mon, 01 Jan 2024 00:00:00 +0000

Multi-Frequency Federated Learning for Human Activity Recognition Using Head-Worn Sensors

Mon, 01 Jan 2024 00:00:00 +0000

BiHeartS

Sun, 01 Jan 2023 00:00:00 +0000

Sleep is the primary mean of recovery from accumulated fatigue and thus plays a crucial role in fostering people’s mental and physical well-being. Sleep quality monitoring systems are often implemented using wearables that leverage their sensing capabilities to provide sleep behaviour insights and recommendations to users. Building models to estimate sleep quality from sensor data is a challenging task, due to the variability of both physiological data, perception of sleep quality, and the daily routine across users. This challenge gauges the need for a comprehensive dataset that includes information about the daily behaviour of users, physiological signals as well as the perceived sleep quality. In this paper, we try to narrow this gap by proposing Bilateral Heart rate from multiple devices and body positions for Sleep measurement (BiHeartS) dataset. The dataset is collected in the wild from 10 participants for 30 consecutive nights. Both research-grade and commercial wearable devices are included in the data collection campaign. Also, comprehensive self-reports are collected about the sleep quality and the daily routine.

BiHeartS: Bilateral Heart Rate from multiple devices and body positions for Sleep measurement Dataset

Sun, 01 Jan 2023 00:00:00 +0000

Enhancing XGBoost with Heuristic Smoothing for Transportation Mode and Activity Recognition

Sun, 01 Jan 2023 00:00:00 +0000

HeartS

Sun, 01 Jan 2023 00:00:00 +0000

Recognition of Engagement from Electrodermal Activity Data Across Different Contexts

Sun, 01 Jan 2023 00:00:00 +0000

Prof. Silvia Santini’s contribution on La Regione

Mon, 21 Nov 2022 00:00:00 +0000

Professor Silvia Santini discussed how digital devices and apps are increasingly integrated into daily life and how this creates both concerns and opportunities for well-being.

Source: .

Our group members ranked among the world’s top 2% of scientists

Wed, 26 Oct 2022 00:00:00 +0000

Marc Langheinrich and Martin Gjoreski were included in Stanford University and Elsevier science-wide citation indicator lists for top scientists.

Source: .

Daniel Gatica-Perez presents Towards Increasing Diversity in Mobile Sensing Research

Tue, 18 Oct 2022 00:00:00 +0000

Professor Silvia Santini hosted Daniel Gatica-Perez from EPFL for a seminar on diversity and mobile sensing research.

Source: .

M2Sleep

Fri, 01 Apr 2022 00:00:00 +0000

Personal informatics systems can help people promote their health and well-being. Recent studies have shown that such systems can be used to infer relevant health indicators such as, e.g., stress, anxiety, and sleeping habits. While automatic detection of sleep has been studied extensively, there is a lack of studies exploring how population and personalized models influence the performance of sleep detection. In this article, we address this challenge by investigating the recognition of sleep/wake stages and high/low sleep quality with a focus on the impact of personalized models. To evaluate our approach, we collect a dataset of physiological signals and self-reports about sleep/wake times and sleep quality score. The dataset contains 6557 hours of sensor data collected using wristbands from 16 participants over one month. Our results show that personalized models perform significantly better than population models for sleep quality recognition, and are comparably good for sleep stage detection. The balanced accuracy for sleep/wake and high/low sleep quality are 92.2% and 61.51%, which are significantly higher than baseline classifiers.

Prof. Marc Langheinrich’s contribution on La Regione

Mon, 21 Feb 2022 00:00:00 +0000

Professor Marc Langheinrich discussed digital privacy and the evolution of personal data handling in a feature with La Regione.

Source: .

Handling Missing Data For Sleep Monitoring Systems

Sat, 01 Jan 2022 00:00:00 +0000

Heart Rate During Sleep Measured Using Finger-, Wrist- and Chest-Worn Devices: A Comparison Study

Sat, 01 Jan 2022 00:00:00 +0000

Multi-task Learning for Stress Recognition

Sat, 01 Jan 2022 00:00:00 +0000

On the Impact of Lateralization in Physiological Signals from Wearable Sensors

Sat, 01 Jan 2022 00:00:00 +0000

On the mismatch between measured and perceived sleep quality

Sat, 01 Jan 2022 00:00:00 +0000

Software-Based Remote Network Attestation

Sat, 01 Jan 2022 00:00:00 +0000

The Role of Model Personalization for Sleep Stage and Sleep Quality Recognition Using Wearables

Sat, 01 Jan 2022 00:00:00 +0000

HAFAR

Mon, 18 Oct 2021 00:00:00 +0000

Head gestures and facial expressions – like, e.g., nodding or smiling – are important indicators of the quality of human interactions in physical meetings as well as in computer-mediated settings. Computer systems able to recognize such behavioral cues can support and improve human interactions. Several researchers have thus tackled the problem of automatically recognizing head gestures and facial expressions, mainly leveraging video data. In this paper, we instead consider inertial signals collected from unobtrusive, ear-mounted devices. We focus on typical activities performed during social interactions – head shaking, nodding, smiling, talking and yawning – and propose a hierarchical classification approach to discriminate them from each other. Further, we investigate whether the transfer of knowledge learned from publicly available datasets leads to further performance improvements. Our results show that the combined use of our hierarchical approach and transfer learning allows the classifier to discriminate head and mouth activities with an F1 score of 84.79, smile, talk and yawn with an F1 score of 45.42, and nodding and head shaking with an F1 score of 88.24, outperforming shallow classifiers by 2-9 percentage points.

Tutorial at the Neuroergonomics 2021 conference in Munich

Mon, 23 Aug 2021 00:00:00 +0000

Martin Gjoreski co-organized a tutorial on machine learning for cognitive load inference from physiological signals at Neuroergonomics 2021.

Source: .

Automatic Recognition of Flow During Work Activities Using Context and Physiological Signals

Fri, 01 Jan 2021 00:00:00 +0000

Biometric recognition using wearable devices in real-life settings

Fri, 01 Jan 2021 00:00:00 +0000

Detecting Verbal and Non-Verbal Gestures Using Earables

Fri, 01 Jan 2021 00:00:00 +0000

Hierarchical Classification and Transfer Learning to Recognize Head Gestures and Facial Expressions Using Earbuds

Fri, 01 Jan 2021 00:00:00 +0000

Synchronous Transmissions in Low-Power Wireless: A Survey of Communication Protocols and Network Services

Fri, 01 Jan 2021 00:00:00 +0000

EDArtifact

Mon, 15 Jun 2020 00:00:00 +0000

Recent wearable devices enable continuous and unobtrusive monitoring of human’s physiological parameters, like e.g., electrodermal activity and heart rate, over long periods of time in everyday life settings. Continuous monitoring of these parameters enables the creation of systems able to predict affective states and stress with the goal of providing feedback to improve them. Deployment of such systems in everyday life settings is still complex and prone to errors due to the low quality of the collected data impacted by the presence of artifacts. In this paper we present an automatic approach to detect artifacts in electrodermal activity (EDA) signals collected in-the-wild over long periods of time. To this end we first perform a systematic literature review and compile a set of guidelines for human annotators to label artifacts manually and we use these labels as ground-truth to test our automatic approach. To evaluate our approach, we collect physiological data from 13 participants in-the-wild and two human annotators label 107.56 hours of this data set. We make the data set publicly available to other researchers upon request. Our model achieves a recall of 98% for clean and shape artifacts classification on data collected in-the-wild using leave-one-subject-out cross-validation, which is 42 percentage points higher than the baseline. We show that state of the art approaches do not generalize well when tested with completely in-the-wild data and identify only 17% of the artifacts present in our data set, even after manual adaption. We further test the robustness of our approach over time using leave-one-day-out and achieve very similar performance. We then introduce a new metric to evaluate the quality of EDA segments that considers the impact of not only artifacts in the shape of EDA but also artifacts generated by environmental temperature changes or user’s high intensity movement. Our results imply that we can eliminate the need for human annotators or significantly reduce the time they need to label data. Also, our approach can be used in an online manner to automatically detect artifacts in EDA signals.

A Multi-Sensor Approach to Automatically Recognize Breaks and Work Activities of Knowledge Workers in Academia

Wed, 01 Jan 2020 00:00:00 +0000

Detection of Artifacts in Ambulatory Electrodermal Activity Data

Wed, 01 Jan 2020 00:00:00 +0000

Synchronous Transmissions in Low-Power Wireless: A Survey of Communication Protocols and Network Services

Wed, 01 Jan 2020 00:00:00 +0000

USILaughs

Mon, 20 May 2019 00:00:00 +0000

A growing number of pervasive systems integrate emotion recognition capabilities with the aim of fostering and promoting human health and wellbeing. In this paper, we argue that enhancing these systems with the ability to detect laughter episodes automatically would improve their effectiveness. This is because laughter is one of the most expressive behavioral cues for positive emotions. However, the existing approaches for laughter recognition rely on the use of obtrusive devices or of video and audio cues and have thus limited applicability in real-world settings. To overcome this limitation, we evaluate the feasibility of a novel, multi-modal approach to recognize laughter episodes using physiological and body movement data gathered with unobtrusive, wrist-worn devices. To assess the performance of our method, we collect an extensive data set of laughter episodes, which we also make publicly available. Our results show that laughter episodes can be distinguished from non-laughter episodes with an accuracy of 81%. Further, we demonstrate that the signatures left by laughter episodes on physiological and body-movement data differ significantly from those caused by slightly intense motions or cognitive load tasks.

Energy-Aware Coflow and Antenna Scheduling for Hybrid Server-Centric Data Center Networks

Tue, 01 Jan 2019 00:00:00 +0000

Evaluating the accuracy of heart rate sensors based on photoplethysmography for in-the-wild analysis

Tue, 01 Jan 2019 00:00:00 +0000

INFAS: In-Network Flow mAnagement Scheme for SDN Control Plane Protection

Tue, 01 Jan 2019 00:00:00 +0000

Laughter Recognition Using Non-invasive Wearable Devices

Tue, 01 Jan 2019 00:00:00 +0000

Using Unobtrusive Wearable Sensors to Measure the Physiological Synchrony Between Presenters and Audience Members

Tue, 01 Jan 2019 00:00:00 +0000

Whisper: Fast Flooding for Low-Power Wireless Networks

Tue, 01 Jan 2019 00:00:00 +0000

6th DACH+ Conference on Energy informatics (EnInf 2017)

Mon, 01 Jan 2018 00:00:00 +0000

Energy-efficient SDN control and visualization

Mon, 01 Jan 2018 00:00:00 +0000

REMO: Resource efficient distributed network monitoring

Mon, 01 Jan 2018 00:00:00 +0000

Selecting Individual and Population Models for Predicting Human Mobility

Mon, 01 Jan 2018 00:00:00 +0000

Unobtrusive Assessment of Students' Emotional Engagement during Lectures Using Electrodermal Activity Sensors

Mon, 01 Jan 2018 00:00:00 +0000

Using Students' Physiological Synchrony to Quantify the Classroom Emotional Climate

Mon, 01 Jan 2018 00:00:00 +0000

Whisper: Fast Flooding for Low-Power Wireless Networks

Mon, 01 Jan 2018 00:00:00 +0000

Editorial

Sun, 01 Jan 2017 00:00:00 +0000

Energy-aware coflow and antenna scheduling for hybrid server-centric data center networks

Sun, 01 Jan 2017 00:00:00 +0000

Every Byte Counts: Selective Prefetching for Mobile Applications

Sun, 01 Jan 2017 00:00:00 +0000

In-classroom self-tracking for teachers and students: preliminary findings from a pilot study

Sun, 01 Jan 2017 00:00:00 +0000

Keep the Beat: On-The-Fly Clock Offset Compensation for Synchronous Transmissions in Low-Power Networks

Sun, 01 Jan 2017 00:00:00 +0000

Secure and energy-efficient interconnects for board-to-board communication

Sun, 01 Jan 2017 00:00:00 +0000

Concurrent Transmissions for Communication Protocols in the Internet of Things

Fri, 01 Jan 2016 00:00:00 +0000

Quantifying the Uncertainty of Next-Place Predictions

Fri, 01 Jan 2016 00:00:00 +0000

A Metric for Adaptive Routing on Trustworthy Paths

Thu, 01 Jan 2015 00:00:00 +0000

A population model for predicting human mobility

Thu, 01 Jan 2015 00:00:00 +0000

A step towards a protocol-independent measurement framework for dynamic networks

Thu, 01 Jan 2015 00:00:00 +0000

An analysis of the YouNow live streaming platform

Thu, 01 Jan 2015 00:00:00 +0000

Automated customer segmentation based on smart meter data with temperature and daylight sensitivity

Thu, 01 Jan 2015 00:00:00 +0000

Household occupancy monitoring using electricity meters

Thu, 01 Jan 2015 00:00:00 +0000

The UbiDCC Data Set: Collecting Wi-Fi and Bluetooth Scans During a Large-scale Conference

Thu, 01 Jan 2015 00:00:00 +0000

Towards enabling concurrent transmissions in heterogeneous networks

Thu, 01 Jan 2015 00:00:00 +0000

How the availability of Wi-Fi connections influences the use of mobile devices

Wed, 01 Jan 2014 00:00:00 +0000

Locator: a self-adaptive framework for the recognition of relevant places

Wed, 01 Jan 2014 00:00:00 +0000

Smart heating control with occupancy prediction: how much can one save?

Wed, 01 Jan 2014 00:00:00 +0000

The ECO data set and the performance of non-intrusive load monitoring algorithms

Wed, 01 Jan 2014 00:00:00 +0000

A site properties assessment framework for wireless sensor networks