AI-Directed XR User Behavior Logs from an Industrial Heritage Museum Study

37 participants' multimodal telemetry data from a four-scene XR experience at the santralistanbul Energy Museum. The dataset includes logged movement, head orientation, and hand movements used to classify users into one of four interaction profiles. It was collected by Başak Çakmak in 2026 to evaluate a curatorially constrained, profile-aware procedural content generation framework.

Use Cases

• Training user profile classification models based on real-time movement and orientation data.
• Studying correlations between telemetry-derived profiles and self-reported user types.
• Analyzing the relationship between immersion, usability, and heritage perception change scores.
• Benchmarking rule-based expert systems with probabilistic scoring for personalized XR content.

Strengths

• Includes ground-truth validation with a reported 83.8% tie-aware accuracy between telemetry and self-reported profiles.
• Contains pre/post questionnaire data showing a statistically significant change in heritage perception (mean paired change = 1.35, p<.001).
• Data collection was multimodal, capturing movement, head orientation, and hand movements.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The exploratory sample size of 37 participants may limit generalizability.

Provenance

Source

Başak Çakmak via figshare.

Collection Method

Collected via a multimodal logging system during a single-group deployment study of a standalone XR headset experience.

Freshness

Last updated 2026-06-02 04:21:13; freshness should be verified.

Geography

Data collected at the santralistanbul Energy Museum.