Dark Mode
iPhone Accelerometer and Gyroscope Motion Data
Product Reviews & Feedback
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
Sensor data collected from an iPhone 5c's accelerometer and gyroscope is used to differentiate between walking and running activities. Smartphones increasingly use sensors to track fitness activities like daily footsteps, and this dataset provides the raw data to build more advanced models. The objective is to use this data to build, test, and compare classification models capable of accurately predicting whether a person is walking or running, a feature that could be incorporated into future IOS devices or fitness applications.
Columns
- date: The date the measurement was recorded.
- time: The exact time the measurement was recorded.
- username: The name of the user who contributed the data.
- wrist: The wrist on which the device was worn. Coded as "0" for the left wrist and "1" for the right wrist.
- activity: The type of activity being performed. This is the target label, coded as "0" for walking and "1" for running.
- acceleration_x: Sensor data representing acceleration along the x-axis.
- acceleration_y: Sensor data representing acceleration along the y-axis.
- acceleration_z: Sensor data representing acceleration along the z-axis.
- gyro_x: Sensor data from the gyroscope along the x-axis.
- gyro_y: Sensor data from the gyroscope along the y-axis.
- gyro_z: Sensor data from the gyroscope along the z-axis.
Distribution
The dataset is provided in a single CSV file named
Kinematics_Data.csv, with a file size of 7.5 MB. It contains 88,588 sensor data samples, structured across 11 columns. The data was collected at a frequency of approximately 5.4 samples per second in 10-second intervals.Usage
Ideal applications for this dataset include developing and training machine learning models for human activity recognition. It is particularly useful for building binary classifiers to distinguish between walking and running. The data can be used to experiment with different classification algorithms, perform hyperparameter tuning, and compare model evaluation metrics. It also serves as a practical dataset for data cleaning and preparation exercises.
Coverage
The data was collected over a specific time range in 2017, from 30 June to 17 July. The sources do not specify the geographic location where the data was collected. All data points were contributed by a single user.
License
CC0: Public Domain
Who Can Use It
- Data Scientists and Machine Learning Engineers: Can use this data to build and fine-tune classification models for fitness and health applications.
- Students and Researchers: Can use this dataset for academic projects related to sensor data analysis, kinematics, and mobile computing.
- App Developers: Can use insights from models built on this data to develop features for fitness tracking applications on mobile devices.
Dataset Name Suggestions
- iOS Device Kinematics: Walking vs. Running
- Human Activity Recognition: Sensor Data
- iPhone Accelerometer and Gyroscope Motion Data
- Wearable Sensor Data for Activity Classification
Attributes
Original Data Source: iPhone Accelerometer and Gyroscope Motion Data
QUALITY