Code Book

Code Book:

Here's a quick quide to the data structure so people can understand what's going on - I hope :)

The data has many parts but is ultimately a set of records composed of a test and training set.
In the main directory, feature, activity and README files provide more descriptive information 
relating to the data and column names used in the data.

In merging the data:
  - I decided to rind X_test & X_train data followed by cbind'ing the subject data.
  - I later add the activities once the IDs have been converted to names using the activity_labels file.

To extract mean & standard deviation info from the data:
  - I decided to use ony data having a mean() or std() in the colname.
  - I left out other instances e.g. meanFreq, as I felt it wasn't a true calculated mean of values.

In naming the variables in the final data, I used the following methodology:
  - Names starting with a lowercase 'f' were renamed with a starting Frequency.
  - Names starting with lowercase 't' were renamed starting with Time.
  - General tidying included full names instead of short ones, capitalising terms & removing dulicates.
  - I removed all unwanted chars "-", "(" & ")"
  - I Attached a literal meaning to the X,Y & Z axis terms.
  - Finally I removed leading and trailing whitespace & placed spaces between all terms to have it like a sentence.