dobrinpetkov / getting_and_cleaning_data Goto Github PK

#Getting and Cleaning Data - Course Project

The script assumes that there is folder datasets with the Samsung data in the working directory. Afterwards, the script follows the assignment requirements.

1. Read train and test datasets - the following datasets are read:
   • activity_labels.txt
   • features.txt
   • test/subject_test.txt
   • test/X_test.txt
   • test/y_test.txt
   • train/subject_train.txt
   • train/X_train.txt
   • train/y_train.txt

The datasets in test/Inertial Signals/ and train/Inertial Signals/ are not read as we do not need them for the analysis.

2. Steps 2 - extracting measurements on the mean and standard deviation, merging the test and train datasets and naming the variables with more descriptive names. Here I have assumed that by measurements on the mean and standard deviation for each measurement it is meant columns that contain one of the strigns mean() or std(). For example, that implies that fBodyAcc-mean()-X is included and fBodyAcc-meanFreq()-X is not as the latter is weighted average of the frequency components. Merging is done via rbind() because it is simple and it does not reorder the rows.

3. Step 3, which is adding descriptive activity names to name the activities in the dataset, is done next.

4. The last step is creating new tidy data set with the average of each variable for each activity and each subject, and saving that data set in output.txt.