getdata-CourseProject - Explanation R script runAnalysis.R

Precondition

All files have to be in same folder - your 'working directory'. Since the names of the files are unique, that won't cause a problem.

File List

• R script: 'run_Analysis.R'
• List of all features: 'features.txt'
• List of all activities with class labels: 'activity_labels.txt'
• Subjects test set: 'subject_test.txt'
• Test set: 'X_test.txt'
• Test labels 'y_test.txt'
• Subjects train set: 'subject_train.txt'
• Training set: 'X_train.txt'
• Training labels: 'y_train.txt'

STEP 1

##Task: Merge the training and the test sets to create one data set.

###Column Names

• read the features.txt as vector

###Test Data

• read the test set
• add the column names from features.txt
• read the test labels
• add column name "Activities"
• read the subjects test set
• add column name "Subject"
• combine the three files in tow steps

###Train Data

• read the train set
• add the column names from features.txt
• read the train labels
• add column name "Activities"
• read the subjects train set
• add column name "Subject"
• combine the three files in tow steps via 'cbind()'

Complete Data

• combine test and train data via 'rbind()' to one data set

STEP 2

##Task: Extract only the measurements on the mean and standard deviation for each measurement.

###Identify Columns

• create a vector with column numbers for all measurements with "mean" oder "std" in the variable name
• 79 results: 33x "std" and 46x "mean"

###Subset Data

• extract the measurements
• including the columns "Subject" and "Activities"

STEP 3

Task: Use descriptive activity names to name the activities in the data set.

###Column Names

• read the activity list including class labels

###Merge

• merge the activity list with the data set
• with the order (first activity list, then data set) and the argument 'all=TRUE' every appearance of the corresponding number is considered

###CleanUp

• delete the column with the activity numbers
• replace the column name "V2" with the more descriptive "Activities"

STEP 4

##Task: Appropriately label the data set with descriptive variable names.

• general structure of the variables: DESCRIPTION- STD or MEAN - AXIS
• the labels should not be to long for a better readability (abbreviations like "Acc" for Accelaration are being used)
• for this reason the labels also contains up and lower case (camel case)
• see CODEBOOK.md for detailed informationen about the variable and their names

Adjustments to the variable names

• "BodyBody" to "Body"
• "tBody" to "time_Body"
• "fBody" to "freq_Body"

STEP 5

##Task: Create a second, independent tidy data set with the average of each variable for each activity and each subject.

###Reshape

• reshape the data set with 'melt()' produces a tall, skinny dataset
• produce the final dataset with the mean values via 'dcast()'

###Output

• the final data "data_final.txt" set has for every 30 subject each with 6 activities a row (180 rows) a mean value for each mean and std variable
• ordered by "Activities" then by "Subject"
• the data set is space delimited, can be read with 'read.table'

###Tidy Dataset

1. Each variable forms a column
2. Each observation forms a row
3. Data is be saved in one file per table
4. A row at the top of the file with variable names is included.
5. Variable names are human readable.