README

Summary

This document summarizes the steps used to generate and analyze a cleaned data set for the wearable computing data provided by UCI.

Dataset

The analysis assumes that you have downloaded and unzipped the following file into your working directory: Data Set

More information about this dataset can be found here: Human Activity Recognition Using Smartphones Data Set

Requirements

R Libraries

The following R libraries were used in this analysis

• dplyr - To work with data tables
• stringr - For performing column replacement to clean up final output

Dataset Files

For this Analysis, the following files in the ZIP archive were used:

• /train/subject_train.txt - Identifies the subject ID for each row of data in X_train and Y_train files.
• /train/X_train.txt - Each row contains 561 columns of data corresponding to a experiment.
• /train/y_train.txt - Identifies which activity was being performed (sitting, standing, etc.). Labels reside in activity_labels.txt.
• /test/subject_test.txt - Identifies the subject ID for each row of data in X_test and Y_test files.
• /test/X_test.txt - Each row contains 561 columns of data corresponding to a experiment.
• /test/y_test.txt - Identifies which activity was being performed (sitting, standing, etc.). Labels reside in activity_labels.txt.
• /features.txt - Labels for the 561 columns in the X_train and X_test files.
• /activity_labels.txt - Activity labels (1 - WALKING, 2 - WALKING_UPSTAIRS, etc.)

Analysis Script

The analysis steps below are performed in run_analysis.R. Running this file with the necessary data set in the working directory will generate a text file 'SamsungSummaryData.txt' which summarizes the mean and standard deviation columns for each combination of Subject and Activity.

##Analysis Steps See Codebook.md