Introduction

The script run_analysis.Rperforms the 5 steps described in the README.md

First, read data using read.table(),the datasets we need is:

1.x_train.txt/x_test.txt: which include numeric datas of varible features

2.y_train.txt/y_test.txt: which include interger flags of activities of every features above

3.subject_x.txt/subject_y.txt: which includes interger flags of subjects

4.features.txt: which includes feature names of every feature types

5.activity_labels.txt: which describes the relationship between interger flags and activity types
Merge train and test datasets using rbind(), so we get one features data frame, one activity data frame, and one subject data frame.
Merge all three data frames using cbind().
Then, select only those columns with the mean and standard deviation measures to get a subset. After extracting these columns, they are given the correct names, taken from features.txt.
We use for() loop for taking the activity names from activity_labels.txt, then give those names to the 'activity' column.
On the whole subset, all columns with vague column names are corrected.
Finally, we generate a new dataset with all the average measures for each subject and activity, and output it to a .txt file.

Provide feedback