Introduction

The purpose of this project is to demonstrate the ability to collect, work with, and clean a data set. Please refer to the CodeBook.md for more details about input data sets and their variables.

The script run_analysis.R is logicaly devided into 6 steps (Step 0 - Step 5). Below you can find a detailed description of how all of the scripts work.

Step 0

Sets working directory

Note! In order to avoid any errors you need to change the path to UCI HAR Dataset folder according to the location on your computer

setwd("C:\\Users\\ekaraseva\\Desktop\\data\\UCI HAR Dataset")

Reads files required for analysis

activity_labels<-read.table("activity_labels.txt")
features<-read.table("features.txt")
subject_train<-read.table("./train/subject_train.txt")
X_train<-read.table("./train/X_train.txt")
y_train<-read.table("./train/y_train.txt")
subject_test<-read.table("./test/subject_test.txt")
X_test<-read.table("./test/X_test.txt")
y_test<-read.table("./test/y_test.txt")`

Step 1

Merges the training and the test sets to create one data set

1. 'train/X_train.txt': Training set and 'train/y_train.txt': Training labels are merged together in order to identify types of activities tracked
2. 'test/X_test.txt': Test set and 'test/y_test.txt': Test labels are merged together in order to identify types of activities tracked
3. Data Sets from points 1 and 2 above were binded together in order to create one data set test_and_train_df

Renames variables in test and train data sets according to the list of all features

names(X_test)<-features[,2]
names(X_train)<-features[,2]
names(subject_test)<-"subject"
names(subject_train)<-"subject"
names(y_test)<-"activity"
names(y_train)<-"activity"

Binds data sets with the lables

test_df<-cbind.data.frame(subject_test, y_test, X_test)
train_df<-cbind.data.frame(subject_train, y_train, X_train)
test_and_train_df<-rbind.data.frame(test_df, train_df)

Step 2

Extracts only the measurements on the mean and standard deviation for each measurement

Regular expressions are used within grep FUN to identify required variable, i.e. ones that included mean() and std() obesrvations. Returned vector was applied to test_and_train_df to filter out other variables. The resulting data frame named as filtered_full_df1.

Locates column numbers that contain mean or standard deviation measurements by exact match to mean() and std()

required_columns_vector<-grep("subject|activity|\\bmean()\\b|\\bstd()\\b", names(test_and_train_df),ignore.case = TRUE)

Creates a filterted data frame that contains only required columns/measurments

filtered_full_df1<-test_and_train_df[,required_columns_vector]

Step 3

Uses descriptive activity names to name the activities in the data set

filtered_full_df1 is merged with activity_labels in order to match numbers with activity names. The names were populated but the numbers were dropped, resulting in creation of another intermediate data frame called human_activities_full.

library(dplyr)
human_activities_full <- filtered_full_df1 %>%
                          merge(activity_labels,by.x="activity",by.y="V1",all=TRUE) %>%
                          select(2,69,3:68) %>%
                          rename(activity=V2) %>%
                          arrange(subject,activity)

Step 4

Appropriately labels the data set with descriptive variable names

A series of editing actions was applied to variable names in human_activities_full. Included: elaboration on abbriviated terms, getting rid of special symbols like "-" and "(), lowering all cases.

NewVarNames <- gsub("-","",names(human_activities_full))
NewVarNames <- gsub("^t","timesignalof",NewVarNames)
NewVarNames <- gsub("^f","frequencysignalof",NewVarNames)
NewVarNames <- gsub("[()]","",NewVarNames)
NewVarNames <- gsub("mean","average",NewVarNames)
NewVarNames <- gsub("std","standarddeviation",NewVarNames)
NewVarNames <- gsub("Body","bodymotion",NewVarNames)
NewVarNames <- gsub("Gravity","gravitationalmotion",NewVarNames)
NewVarNames <- gsub("Acc","fromaccelerometer",NewVarNames)
NewVarNames <- gsub("Gyro","fromgyroscope",NewVarNames)
NewVarNames <- gsub("Mag","magnitude",NewVarNames)
NewVarNames <- tolower (NewVarNames)
names(human_activities_full)<-NewVarNames

Step 5

From the data set in step 4, creates a second, independent tidy data set with the average of each variable for each activity and each subject

human_activities_full has been grouped by activity types. The data frame has been summarized by activity types using mean FUN. The resulting and final data frame was called human_activities_summary.

human_activities_summary<- human_activities_full %>% 
              group_by(subject, activity) %>%
              summarise_all(funs(mean(., na.rm=TRUE))) %>%
              as.data.frame()

print(human_activities_summary)
#or
write.table(human_activities_summary,file="run_analysis.txt",row.names = FALSE)