Calling libraries

library(dplyr)

## Warning: package 'dplyr' was built under R version 3.3.2

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(tidyr)

## Warning: package 'tidyr' was built under R version 3.3.2

Downloading and unzipping files

fileurl <- "https://d396qusza40orc.cloudfront.net/getdata%2Fprojectfiles%2FUCI%20HAR%20Dataset.zip"

download.file(fileurl,destfile = "./UCI HAR Dataset.zip", mode = "wb")

unzip("./UCI HAR Dataset.zip")

Reading files into tables

X_test <- read.table("./UCI HAR Dataset/test/X_test.txt")
X_train <- read.table("./UCI HAR Dataset/train/X_train.txt")

Y_test <- read.table("./UCI HAR Dataset/test/Y_test.txt")
Y_train <- read.table("./UCI HAR Dataset/train/Y_train.txt")

subject_test <- read.table("./UCI HAR Dataset/test/subject_test.txt")
subject_train <- read.table("./UCI HAR Dataset/train/subject_train.txt")

features <- read.table("./UCI HAR Dataset/features.txt")
activity_labels <- read.table("./UCI HAR Dataset/activity_labels.txt")

Combining train and test tables of each content set. Converting to table format

X <- rbind(X_test, X_train);                    X <- tbl_df(X)
Y <- rbind(Y_test, Y_train);                    Y <- tbl_df(Y)

subject <- rbind(subject_test, subject_train);  subject <- tbl_df(subject)

features <- tbl_df(features)
activity_labels <- tbl_df(activity_labels)

Removing redundant tables

rm(X_test, X_train, Y_test, Y_train, subject_test, subject_train)

Setting Variable names to features' descriptions

column_names <- sapply(features$V2, as.character)
names(X) <- column_names

Extracting only the measurements on the mean and standard deviation for each measurement.

subset_X <- X[,grep("mean|std", names(X))]
        subset_X <- subset_X[,-grep("meanFreq()|Jerk|Mag|^f",names(subset_X))]

Here we filter out measures which are not raw measurements, but rather derived Such as Jerk, Magnitude, Furier Transformation and Mean Frequency

Making variable names readable

names(subset_X) <- sub("Acc", "Acceleration", names(subset_X))
names(subset_X) <- sub("Gyro", "Gyroscope", names(subset_X))
names(subset_X) <- sub("^t", "Time", names(subset_X))
names(subset_X) <- sub("-", "Signal", names(subset_X))
names(subset_X) <- sub("\\()", "", names(subset_X))

Using activity Names

Y <- Y %>% inner_join(activity_labels, by = "V1") %>%
        transmute(activity = V2)

Subject variable name change from "V1" to "subject"

names(subject) <- "subject"

Combining subject, subset_X, and Y tables

df <- subject %>% cbind(Y) %>% cbind(subset_X) %>% tbl_df()

Making data narrow to perform averaging across subject, activity, and measure

tidy <- df %>% gather(measure, value, 3:19)
tidy <- tidy %>% group_by(subject, activity, measure) %>%
        summarize(meanvalue = mean(value))

Reverting data back to the wide format when in each column there is a different measurement mean

final_table <- tidy %>% spread(measure, meanvalue)