Artificially introduced MCAR dataset for missing data analysis in activity recognition

This dataset was generated by introducing Missing Completely at Random (MCAR) data into activity recognition measurements collected from a smart bracelet. The dataset simulates real-world missing data scenarios and is useful for evaluating imputation methods and their impact on activity classification and health monitoring accuracy. Data was originally recorded during various activities, including walking, running, climbing stairs, and resting, with artificial missing values introduced to simulate MCAR conditions.