Repository contains Coursera Getting nad Cleaning Data Course project.

The main script in this project is 'run_analysis.R' which performs the following tasks:

1. Downloads data from https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip
2. Unzips all files into "./data" directory
3. Reads all the data for Subject, X, and Y, both training and test datasets using read.csv function.
4. Uses rbind to merge training and test data sets, and 'cbind' to merge Subject, Activity and Measurements.
5. Gives hard-coded names for Subject and Activity columns, reads "features.txt" to give meaningful names for other columns in the combined data set.
6. Uses grepl to find all column names containing "mean" or "std". Columns containing "meanFreq" are excluded since these contain rate of measurement but not sensor data.
7. Reads IDs and names of activities from "activity_labels.txt" and converts activity ids in the dataset to descriptive names of activities using factor.
8. Uses ddply to create a second data set with the average of each variable for each activity and each subject. Data is split by variables Subject and Activity, colMeans are used to calculate means on all columns except first two (Subject and Activity).
9. Uses write.csv to save the final tidy dataset as a CSV file names "./data/tidydata.txt".