/gettingandcleaning

Primary LanguageR

Human Activity Recognition using Smartphones Data Analysis

The gettingandcleaning repository contains the below files expected by the project:

File Name Description
README.md Readme file that explains each file in the repository and a brief description of the file
run_analysis.R R Function file that creates the tidy data set
CodeBook.md Contains step by step explanation of the run_analysis function and the desc of the variable names in the final output file final_tidy_data.txt
final_tidy_data.txt Tidy data set produced by running run_analysis()
final_labeled_data.txt Text file produced before the last step of producing tidy data

The repository also contains the below data files that was supplied for the project:

Data File Description
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
X_train.txt Training set
y_train.txt Training labels
subject_train.txt Each row identifies the subject who performed the train activity for each window sample
X_test.txt Test set
y_test.txt Test labels
subject_test.txt Each row identifies the subject who performed the test activity for each window sample

Explanation of run_analysis.R

The function run_analysis accepts the working directory workingdir parameter as an input. The parameter has the current directory as the default value.

  1. It sets the working directory
  2. Loads the data.table library
  3. Reads the activity_lablels.txt file into activity_labels data table and names the two columns as activity_id and activity_name
  4. Read the features.txt file into features data table and names the two columns as feature_id and feature_desc
  5. Read the X test and train data and combine them into a Xcombined data table. Name the columns in the data table from the feature_desc column in the features data table
  6. Select only the columns that are mean or standard deviation measurements from the Xcombined data table by searching for mean or std pattern in the column names. Name the resulting set as meanstd
  7. Read the y test and train data and combine them into ycombined data table. Name the column in the data table as activity_id
  8. Read the subject test and train data and combine them into subjectcombined data table. Name the column in the data table as subject_id
  9. Combine the meanstd, ycombined and subjectcombined into a single data table called cd
  10. Merge cd and activity_labels on activity_id
  11. Remove the activity_id column as we have the activity_name and name it as cdt
  12. Write cdt to a file called final_labeled_data.txt with TAB as delimiter
  13. Create a new tidy data set by activity_name and subject and mean of the other variables with TAB as delimiter. Call this the `final_tidy_data.txt

Running run_analysis from R console should produce two output files that are TAB delimited in the working directory. The files are final_labeled_data.txt and final_tidy_data.txt