Saved in:
Bibliographic Details
Main Authors: Witt, Clemens, Leonhardt, Thiemo, Bergner, Nadine, Grillenberger, Mareen
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.22426
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909712193159168
author Witt, Clemens
Leonhardt, Thiemo
Bergner, Nadine
Grillenberger, Mareen
author_facet Witt, Clemens
Leonhardt, Thiemo
Bergner, Nadine
Grillenberger, Mareen
contents Machine learning models are widely used to support stealth assessment in digital learning environments. Existing approaches typically rely on abstracted gameplay log data, which may overlook subtle behavioral cues linked to learners' cognitive strategies. This paper proposes a multimodal late fusion model that integrates screencast-based visual data and structured in-game action sequences to classify students' problem-solving strategies. In a pilot study with secondary school students (N=149) playing a multitouch educational game, the fusion model outperformed unimodal baseline models, increasing classification accuracy by over 15%. Results highlight the potential of multimodal ML for strategy-sensitive assessment and adaptive support in interactive learning contexts.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22426
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Multimodal Late Fusion Model for Problem-Solving Strategy Classification in a Machine Learning Game
Witt, Clemens
Leonhardt, Thiemo
Bergner, Nadine
Grillenberger, Mareen
Machine Learning
Machine learning models are widely used to support stealth assessment in digital learning environments. Existing approaches typically rely on abstracted gameplay log data, which may overlook subtle behavioral cues linked to learners' cognitive strategies. This paper proposes a multimodal late fusion model that integrates screencast-based visual data and structured in-game action sequences to classify students' problem-solving strategies. In a pilot study with secondary school students (N=149) playing a multitouch educational game, the fusion model outperformed unimodal baseline models, increasing classification accuracy by over 15%. Results highlight the potential of multimodal ML for strategy-sensitive assessment and adaptive support in interactive learning contexts.
title Multimodal Late Fusion Model for Problem-Solving Strategy Classification in a Machine Learning Game
topic Machine Learning
url https://arxiv.org/abs/2507.22426