##Introduction
This file describes the data and its variables and informs in detail the transformations performed on the initial The Human Activity Recognition Using Smartphones Data Set using the run_analysis.R script. There is a Glossary at the start explaning the different labels/features present in the initial data set.
##Dataset Cleanup and Transformations: The Human Activity Recognition Using Smartphones Data Set data set consists of totally 563 features (including Activity ID and Subject ID) and total of 10299 observations (including train and test observations). The features expected to be extracted from this feature set is Mean and Standard Deviation measurements. Given there is a possiblity for ambiguity, the coding allows to have
- mean() and std() measurements - 75 Features
- mean(),meanFreq() and std measurements - 88 Features
This is including the SubjectID and ActivityID. By default I took the latter with 88 features. Then the ActivityID (1 to 6) is replaced by the corresponding activity description WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING, LAYING in the reduced dataset. Subsequently the average of the features per SubjectID per ActivityID is calculated and written to the tidy data set. The dimensions of the final tidy data set is 180 * 88 . 180 comes from 6 activities per 30 subjects (6 * 30) . The tidy data set is written out to a file . By default to final.txt in current working directory.
The functions used are
- mergeData : Merge the training and the test sets to create one data set.
- reduceFeatureSpace : Extract only the measurements on the mean and standard deviation for each measurement.
- changeActivityIdToDesc : Change descriptive activity names to name the activities in the data set.
- changeActivityIdToDesc : Appropriately label the data set with descriptive activity names.
- tidyUpData : Creates a independent tidy data set with the average of each variable for each activity and each subject.
- write.table : Write the tidy data to a file
##Glossary of Input Data Features
| Variable Name | Description |
|---|---|
| SubjectID | Total of 30 subjects referenced by numbers 1 to 30 |
| Activity Name | Total of 6 activities performed by the 30 Subjects. The activities being WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING, LAYING |
| ActivtyID | The 6 activities of subjects mapped to ids 1 to 6 |
| X,Y,Z | 3-axial signals in the X, Y and Z directions. Used at end of the feature names) |
| Mean | Mean value |
| Std | Standard deviation |
| MeanFreq | Weighted average of the frequency components to obtain a mean frequency |
| t | time domain signal. Used at start of the feature names: eg: tAcc-XYZ |
| f | frequenct domain signal. Used at start of the feature names: eg fBodyAcc-XYZ |
| Acc | Accelerometer |
| Gyro | Gyroscope |
| BodyAcc | Body Acceleration Signal (derived from Acc features) |
| GravityAcc | Gravitation Acceleration Signal (derived from Acc features) |
| Jerk | Jerk Signals |
| Mag | Magnitude of the three dimenstional signals |
##Used Features
Here is the list of used mean and standard deviation features
| No | VariableName |
|---|---|
| 1 | SubjectID |
| 2 | ActivityName |
| 3 | tBodyAccMeanX |
| 4 | tBodyAccMeanY |
| 5 | tBodyAccMeanZ |
| 6 | tBodyAccStdX |
| 7 | tBodyAccStdY |
| 8 | tBodyAccStdZ |
| 9 | tGravityAccMeanX |
| 10 | tGravityAccMeanY |
| 11 | tGravityAccMeanZ |
| 12 | tGravityAccStdX |
| 13 | tGravityAccStdY |
| 14 | tGravityAccStdZ |
| 15 | tBodyAccJerkMeanX |
| 16 | tBodyAccJerkMeanY |
| 17 | tBodyAccJerkMeanZ |
| 18 | tBodyAccJerkStdX |
| 19 | tBodyAccJerkStdY |
| 20 | tBodyAccJerkStdZ |
| 21 | tBodyGyroMeanX |
| 22 | tBodyGyroMeanY |
| 23 | tBodyGyroMeanZ |
| 24 | tBodyGyroStdX |
| 25 | tBodyGyroStdY |
| 26 | tBodyGyroStdZ |
| 27 | tBodyGyroJerkMeanX |
| 28 | tBodyGyroJerkMeanY |
| 29 | tBodyGyroJerkMeanZ |
| 30 | tBodyGyroJerkStdX |
| 31 | tBodyGyroJerkStdY |
| 32 | tBodyGyroJerkStdZ |
| 33 | tBodyAccMagMean |
| 34 | tBodyAccMagStd |
| 35 | tGravityAccMagMean |
| 36 | tGravityAccMagStd |
| 37 | tBodyAccJerkMagMean |
| 38 | tBodyAccJerkMagStd |
| 39 | tBodyGyroMagMean |
| 40 | tBodyGyroMagStd |
| 41 | tBodyGyroJerkMagMean |
| 42 | tBodyGyroJerkMagStd |
| 43 | fBodyAccMeanX |
| 44 | fBodyAccMeanY |
| 45 | fBodyAccMeanZ |
| 46 | fBodyAccStdX |
| 47 | fBodyAccStdY |
| 48 | fBodyAccStdZ |
| 49 | fBodyAccMeanFreqX |
| 50 | fBodyAccMeanFreqY |
| 51 | fBodyAccMeanFreqZ |
| 52 | fBodyAccJerkMeanX |
| 53 | fBodyAccJerkMeanY |
| 54 | fBodyAccJerkMeanZ |
| 55 | fBodyAccJerkStdX |
| 56 | fBodyAccJerkStdY |
| 57 | fBodyAccJerkStdZ |
| 58 | fBodyAccJerkMeanFreqX |
| 59 | fBodyAccJerkMeanFreqY |
| 60 | fBodyAccJerkMeanFreqZ |
| 61 | fBodyGyroMeanX |
| 62 | fBodyGyroMeanY |
| 63 | fBodyGyroMeanZ |
| 64 | fBodyGyroStdX |
| 65 | fBodyGyroStdY |
| 66 | fBodyGyroStdZ |
| 67 | fBodyGyroMeanFreqX |
| 68 | fBodyGyroMeanFreqY |
| 69 | fBodyGyroMeanFreqZ |
| 70 | fBodyAccMagMean |
| 71 | fBodyAccMagStd |
| 72 | fBodyAccMagMeanFreq |
| 73 | fBodyBodyAccJerkMagMean |
| 74 | fBodyBodyAccJerkMagStd |
| 75 | fBodyBodyAccJerkMagMeanFreq |
| 76 | fBodyBodyGyroMagMean |
| 77 | fBodyBodyGyroMagStd |
| 78 | fBodyBodyGyroMagMeanFreq |
| 79 | fBodyBodyGyroJerkMagMean |
| 80 | fBodyBodyGyroJerkMagStd |
| 81 | fBodyBodyGyroJerkMagMeanFreq |
| 82 | angletBodyAccMeangravity |
| 83 | angletBodyAccJerkMeangravityMean |
| 84 | angletBodyGyroMeangravityMean |
| 85 | angletBodyGyroJerkMeangravityMean |
| 86 | angleXgravityMean |
| 87 | angleYgravityMean |
| 88 | angleZgravityMean |
##References: