rockgecko / getdata-013-project Goto Github PK

This README file provides information relating to the course project submission for the Getting and Cleaning Data course.

The submission consists of:

• A tidy dataset containing the output of the required processing steps, saved as a text file UCI_HAR_summaryData.txt
• The script run_analysis.R used to perform the required processing
• A file CodeBook.md that describes the output data file format and contents
• This README file, which summarizes the processing steps carried out to produce the output file

To replicate the analysis:

• Download the dataset from https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip.
• Unzip the downloaded zip file into the working directory.
• Ensure the dplyr libary is loaded.
• Source/run the run_analysis.R script

The data has been generated starting from the UCI HAR dataset provided, using the script run_analysis.R. The script assumes the unzipped folder "UCI HAR Dataset" is located in the working directory.

The script carries out the following steps:

1. Read the test and training datasets from file
2. Combine the two datasets into a dataframe called combinedData
3. Read the file features.txt to give a dataframe called labels containing the variable names
4. Using regular expressions, search labels for any names containing the text "mean()" or "std()" and store those row numbers
5. Store a subset of combinedData dataframe containing only the variables corresponding to the selected rows in the dataframe meanAndstdData
6. Label the columns of meanAndstdData with the names of the selected variables
7. Read the subject_test.txt and subject_train.txt files and combined the resulting dataframes to obtain subject ids for each observation
8. Read the y_test.txt and y_train.txt files and combined the resulting dataframes to obtain activity ids for each observation
9. Read the activity_labels.txt file to obtain a look-up table of activity id versus activity name
10. Match the activity id for each observation in meanAndstdData with a corresponding activity name and store the names in a dataframe DescActs
11. Add the DescActs and Subjects dataframes to the left of meanAndstdData to produce a dataframe with observations labelled by Subject and Activity
12. Using the group_by() and summarise_each() functions, calculate the mean values of each variable within each activity and subject group and store these as the dataframe summaryData
13. Write the summaryData dataframe out to the file "UCI_HAR_summaryData.txt" with a header row but no row labels