Titre : Automatic classification of emotions using motion sensors and keystroke dynamics on smartphones 

Auteurs : Nicolas Simonazzi
, Jean-Marc Salotti
, Caroline Dubois
, Philippe Le Goff,  

Revue : Cognitive Engineering 

Numéro : Issue 1 

Volume : 8 

Date : 2025/07/21 

DOI : 10.21494/ISTE.OP.2025.1328 

ISSN : 2517-6978 

Résumé : We present the results of a study on a binary classification of emotions, based on data collected through motion sensors and keystrokes of a smartphone and a connected bracelet. To this end, we developed a mobile application to induce emotions through videos and record user interactions. A specific digital self-assessment system was developed based on the Geneva Emotion Wheel to help participants express their emotions. The sensor recordings were labelled according to participants’ statements and video conditions. A method is thus proposed to process the collected temporal data and automatically classify the valence of the declared emotions using machine learning techniques. We tested a general valence classification using all emotions from all individuals and a personalized classification using a subset of emotions from a single individual. The most promising result was obtained with a personalized model, for which we were able to obtain, on average across all participants, two-thirds of correct valence classification, using fused data from different modalities. 

Éditeur : ISTE OpenScience