Back-of-Device Tap Interactions with Inertial Sensor Data

A standardized dataset of back-of-device tap interactions captured using a Samsung Galaxy Note 8. It includes synchronized accelerometer, gyroscope, and high-frequency accelerometer signals segmented around each tap from five participants performing controlled taps on a 3×3 grid.

Use Cases

• Benchmark tap-recognition algorithms using segmented accelerometer and gyroscope signals.
• Develop sensor-fusion methods for mobile interaction by analyzing synchronized accelerometer and gyroscope data.
• Research human-computer interaction by modeling tap patterns from the 3×3 grid layout.
• Train time-series models for gesture recognition using the high-frequency accelerometer signal.

Strengths

• Data was captured from a specific device model (Samsung Galaxy Note 8) and Android version (9) for consistency.
• Includes synchronized data from multiple inertial sensors (accelerometer, gyroscope, high-frequency accelerometer).
• Taps were performed in a controlled setting on a defined 3×3 grid by five participants.

Limitations

• Small participant pool of only five individuals limits demographic and behavioral diversity.
• Data collection was performed in a single, controlled posture (seated, using right hand and index finger), reducing ecological validity.
• The dataset's size, row count, and file formats are unspecified.

Provenance

Source

Harvard Dataverse

Collection Method

Controlled taps performed by participants on a Samsung Galaxy Note 8, with synchronized sensor data capture.

Freshness

Last updated on 2026-02-17.