guidaeta 0.1.0

Utility tools for handling the guidaeta dataset.

GUIDÆTA

A Versatile Interactions Dataset with extensive Context Information and Metadata

The GUIDÆTA interactions dataset was collected within an online user study in the A+CHIS project on Human-Centered Interactive Adaptive Visual Approaches in High-Quality Health Information, conducted between May 12 and June 23, 2025.

Extensive information on scope, data collection, metadata, preliminary analysis can be found in the respective paper ("GUIDÆTA – A Versatile Interactions Dataset with extensive Context Information and Metadata"), published at Smart Tools and Applications in Graphics (STAG) 2025, an annual international conference organized by the Italian Chapter of the Eurographics association.

The raw dataset is publicly available on OSF: https://osf.io/fhvbm/.

Usage

This repository contains everything necessary for handling the raw dataset. I.e., loading, filtering, resampling, analysis and more. All of which is implemented in observations.py. All tools can be used by just importing this file. A few selected use cases are given in main.py:

import observations


# The path pointing to the root of the data
DATA_ROOT = "../CLT/data"

# load everything from files and initialize
sentences, sessions, users, task_answers = observations.load_data(DATA_ROOT)

# get the cognitive load experienced for task two
task_two_answers = [ta for ta in task_answers if ta.task_no==2]
cl = sum([
	sum(ta.cl.scores)/len(ta.cl.scores) for ta in task_two_answers
])/len(task_two_answers)

# get all sessions associated with the 3rd task
s = [s for ta in task_answers for s in ta.sessions if ta.task_no==3]

Publications

When referring to the dataset please cite:

• S. Lengauer, S.A. von Götz, M.T. Hoesch, F. Steinwidder, M. Tytarenko, M.A. Bedek, T. Schreck, GUIDÆTA - A Versatile Interactions Dataset with extensive Context Information and Metadata, 2025.