Skip to content

Latest commit

 

History

History
142 lines (114 loc) · 9.67 KB

CodeBook.md

File metadata and controls

142 lines (114 loc) · 9.67 KB

run_analysis.r

Getting and Cleaning Data September 2014 David Saint Ruby

Background

This script takes the raw data from the UCI HAR Study and creates two consumable files. This script combines up both the test and training data.

You may refer to project specifics here http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones

The first output file is a cleaned file consisting of the mean and standard deviation. The second file is a summary of averages by the subject and their activity.

Input Files

  • 'features_info.txt': Shows information about the variables used on the feature vector.
  • 'features.txt': List of all features.
  • 'activity_labels.txt': Links the class labels with their activity name.
  • 'train/X_train.txt': Training set.
  • 'train/y_train.txt': Training labels.
  • 'test/X_test.txt': Test set.
  • 'test/y_test.txt': Test labels.
  • 'train/subject_train.txt': Each row identifies the subject who performed the activity for each window sample. Its range is from 1 to 30.

Notes:

  • Features are normalized and bounded within [-1,1].
  • Each feature vector is a row on the text file.
  • 30 Subjects participated.

Processing Steps

The test and training files are ingested. The subject and activity files are then appended to this data. The test and training data sets are then combined. We also appended the activity label to each of the observations for ease of use.

For our first outputfile (UCIHARconsolidated), we cleaned up variable names and subsetted the data to means or standard deviations.

For our second output file, we calculated averages for the first file grouped by subject and activity.

Output Files

Both files are saved with write.table with row.name=FALSE.

UCIHARconsolidated.txt

  • subject 1 to 30, units: int person id 1 to 30
  • activity from y files, 1 to 6, units: int activity id
  • activity_label descriptive activity from activity_labels, units: str activity label
  • timeBodyAccMeanForX original variable name of tBodyAcc-mean()-X units:num normalized from -1 to 1
  • timeBodyAccMeanForY original variable name of tBodyAcc-mean()-Y units:num normalized from -1 to 1
  • timeBodyAccMeanForZ original variable name of tBodyAcc-mean()-Z units:num normalized from -1 to 1
  • timeBodyAccStdForX original variable name of tBodyAcc-std()-X units:num normalized from -1 to 1
  • timeBodyAccStdForY original variable name of tBodyAcc-std()-Y units:num normalized from -1 to 1
  • timeBodyAccStdForZ original variable name of tBodyAcc-std()-Z units:num normalized from -1 to 1
  • timeGravityAccMeanForX original variable name of tGravityAcc-mean()-X units:num normalized from -1 to 1
  • timeGravityAccMeanForY original variable name of tGravityAcc-mean()-Y units:num normalized from -1 to 1
  • timeGravityAccMeanForZ original variable name of tGravityAcc-mean()-Z units:num normalized from -1 to 1
  • timeGravityAccStdForX original variable name of tGravityAcc-std()-X units:num normalized from -1 to 1
  • timeGravityAccStdForY original variable name of tGravityAcc-std()-Y units:num normalized from -1 to 1
  • timeGravityAccStdForZ original variable name of tGravityAcc-std()-Z units:num normalized from -1 to 1
  • timeBodyAccJerkMeanForX original variable name of tBodyAccJerk-mean()-X units:num normalized from -1 to 1
  • timeBodyAccJerkMeanForY original variable name of tBodyAccJerk-mean()-Y units:num normalized from -1 to 1
  • timeBodyAccJerkMeanForZ original variable name of tBodyAccJerk-mean()-Z units:num normalized from -1 to 1
  • timeBodyAccJerkStdForX original variable name of tBodyAccJerk-std()-X units:num normalized from -1 to 1
  • timeBodyAccJerkStdForY original variable name of tBodyAccJerk-std()-Y units:num normalized from -1 to 1
  • timeBodyAccJerkStdForZ original variable name of tBodyAccJerk-std()-Z units:num normalized from -1 to 1
  • timeBodyGyroMeanForX original variable name of tBodyGyro-mean()-X units:num normalized from -1 to 1
  • timeBodyGyroMeanForY original variable name of tBodyGyro-mean()-Y units:num normalized from -1 to 1
  • timeBodyGyroMeanForZ original variable name of tBodyGyro-mean()-Z units:num normalized from -1 to 1
  • timeBodyGyroStdForX original variable name of tBodyGyro-std()-X units:num normalized from -1 to 1
  • timeBodyGyroStdForY original variable name of tBodyGyro-std()-Y units:num normalized from -1 to 1
  • timeBodyGyroStdForZ original variable name of tBodyGyro-std()-Z units:num normalized from -1 to 1
  • timeBodyGyroJerkMeanForX original variable name of tBodyGyroJerk-mean()-X units:num normalized from -1 to 1
  • timeBodyGyroJerkMeanForY original variable name of tBodyGyroJerk-mean()-Y units:num normalized from -1 to 1
  • timeBodyGyroJerkMeanForZ original variable name of tBodyGyroJerk-mean()-Z units:num normalized from -1 to 1
  • timeBodyGyroJerkStdForX original variable name of tBodyGyroJerk-std()-X units:num normalized from -1 to 1
  • timeBodyGyroJerkStdForY original variable name of tBodyGyroJerk-std()-Y units:num normalized from -1 to 1
  • timeBodyGyroJerkStdForZ original variable name of tBodyGyroJerk-std()-Z units:num normalized from -1 to 1
  • timeBodyAccMagMean original variable name of tBodyAccMag-mean() units:num normalized from -1 to 1
  • timeBodyAccMagStd original variable name of tBodyAccMag-std() units:num normalized from -1 to 1
  • timeGravityAccMagMean original variable name of tGravityAccMag-mean() units:num normalized from -1 to 1
  • timeGravityAccMagStd original variable name of tGravityAccMag-std() units:num normalized from -1 to 1
  • timeBodyAccJerkMagMean original variable name of tBodyAccJerkMag-mean() units:num normalized from -1 to 1
  • timeBodyAccJerkMagStd original variable name of tBodyAccJerkMag-std() units:num normalized from -1 to 1
  • timeBodyGyroMagMean original variable name of tBodyGyroMag-mean() units:num normalized from -1 to 1
  • timeBodyGyroMagStd original variable name of tBodyGyroMag-std() units:num normalized from -1 to 1
  • timeBodyGyroJerkMagMean original variable name of tBodyGyroJerkMag-mean() units:num normalized from -1 to 1
  • timeBodyGyroJerkMagStd original variable name of tBodyGyroJerkMag-std() units:num normalized from -1 to 1
  • freqBodyAccMeanForX original variable name of fBodyAcc-mean()-X units:num normalized from -1 to 1
  • freqBodyAccMeanForY original variable name of fBodyAcc-mean()-Y units:num normalized from -1 to 1
  • freqBodyAccMeanForZ original variable name of fBodyAcc-mean()-Z units:num normalized from -1 to 1
  • freqBodyAccStdForX original variable name of fBodyAcc-std()-X units:num normalized from -1 to 1
  • freqBodyAccStdForY original variable name of fBodyAcc-std()-Y units:num normalized from -1 to 1
  • freqBodyAccStdForZ original variable name of fBodyAcc-std()-Z units:num normalized from -1 to 1
  • freqBodyAccJerkMeanForX original variable name of fBodyAccJerk-mean()-X units:num normalized from -1 to 1
  • freqBodyAccJerkMeanForY original variable name of fBodyAccJerk-mean()-Y units:num normalized from -1 to 1
  • freqBodyAccJerkMeanForZ original variable name of fBodyAccJerk-mean()-Z units:num normalized from -1 to 1
  • freqBodyAccJerkStdForX original variable name of fBodyAccJerk-std()-X units:num normalized from -1 to 1
  • freqBodyAccJerkStdForY original variable name of fBodyAccJerk-std()-Y units:num normalized from -1 to 1
  • freqBodyAccJerkStdForZ original variable name of fBodyAccJerk-std()-Z units:num normalized from -1 to 1
  • freqBodyGyroMeanForX original variable name of fBodyGyro-mean()-X units:num normalized from -1 to 1
  • freqBodyGyroMeanForY original variable name of fBodyGyro-mean()-Y units:num normalized from -1 to 1
  • freqBodyGyroMeanForZ original variable name of fBodyGyro-mean()-Z units:num normalized from -1 to 1
  • freqBodyGyroStdForX original variable name of fBodyGyro-std()-X units:num normalized from -1 to 1
  • freqBodyGyroStdForY original variable name of fBodyGyro-std()-Y units:num normalized from -1 to 1
  • freqBodyGyroStdForZ original variable name of fBodyGyro-std()-Z units:num normalized from -1 to 1
  • freqBodyAccMagMean original variable name of fBodyAccMag-mean() units:num normalized from -1 to 1
  • freqBodyAccMagStd original variable name of fBodyAccMag-std() units:num normalized from -1 to 1
  • freqBodyAccJerkMagMean original variable name of fBodyBodyAccJerkMag-mean() units:num normalized from -1 to 1
  • freqBodyAccJerkMagStd original variable name of fBodyBodyAccJerkMag-std() units:num normalized from -1 to 1
  • freqBodyGyroMagMean original variable name of fBodyBodyGyroMag-mean() units:num normalized from -1 to 1
  • freqBodyGyroMagStd original variable name of fBodyBodyGyroMag-std() units:num normalized from -1 to 1
  • freqBodyGyroJerkMagMean original variable name of fBodyBodyGyroJerkMag-mean() units:num normalized from -1 to 1
  • freqBodyGyroJerkMagStd original variable name of fBodyBodyGyroJerkMag-std() units:num normalized from -1 to 1

UCIHARconsolidatedMeans.txt

  • same as above file, except that the columns represent the means for the grouping (subject, activity, activity_label)

We acknoledge the following for the data source: Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra and Jorge L. Reyes-Ortiz. Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine. International Workshop of Ambient Assisted Living (IWAAL 2012). Vitoria-Gasteiz, Spain. Dec 2012