Getting and Cleaning Data - Course Project

This repo was created for the Coursera Getting and Cleaning Data .

It contains 4 files:

• README.md : Basic description of the assignement and the run_analysis.R script
• run_analysis.R : The R script used to generate the tidy data set
• result.txt : The tidy data set itself
• Codebook : Description of the data set

Description of the run_analysis.R script

Prerequisites: the UCI HAR Dataset is available in the Workspace. Alternatively, it can be downloaded and unzipped using the script from lines 5-9 (uncommented).

• Initially all data required for the analysis are read to the memory using the read.table function
• All Test and Training data are merged (dataX, dataSubject, dataY)
• The column names of the dataX data frame are modified according to the features - renamed, removed all but "std" & "mean"
• The column names of the dataX data frame further changed to be "nicer"
• activity number in dataY replaced by activity name
• activities(dataY) merged with the dataX and renamed
• subject(dataSubject) merged with the dataX and renamed
• Mean calculated for the dataX (grouped by subject & activity) and stored in a new tidy dataframe resultDF
• resultDF saved to a "result.txt"

Used data frames:

• data_features: contains all the feature names
• dataXTrain: Training data
• dataXTest: Test data
• dataX: dataXTest merged with dataXTrain
• dataYTest: activity description for the Test
• dataYTrain: activity description for the Train
• dataY: dataYTest merged with dataYTrain
• dataSubjectTest: contains the subject for Test
• dataSubjectTrain: contains the subject for Train
• dataSubject: dataSubjectTest merged with dataSubjectTrain
• resultDF: means calculated for dataX stored in a dataframe

Used variables:

• currentDir: Actual directory
• dataDir: root directory of the UCI HAR Dataset
• featuresToKeep: vector of columns, which are to be kept