README

This project processes the Human Activity Recognition dataset from the UCI Machine Learning Repository into two resulting datasets for further analysis.

Raw Data

The raw data was downloaded from

http://archive.ics.uci.edu/ml/machine-learning-databases/00240/

The description of its contents are available at

http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones

Generated Data

The resulting data are stored in two files

• activity.txt - contains data for 30 subjects performing 6 activities.
  Contents include the means and standard deviations of measurements collected from the accelerometer and gyroscope of a smartphone. For details of the measurements, consult the code book.

• means.txt - summarizes the first dataset, providing mean statistics of all variables for each subject and each activity.

Instructions

To process the dataset with a given working directory DIR

1. extract the raw data into the subdirectory DIR/UCI HAR Dataset
2. execute the R script run_analysis.R

The script contains functions that can be easily customized should your directory structure be different.