This was a project for the Getting and Cleaning Data course in Coursera’s Data Science specialization track. The purpose of the project was to practice cleaning and merging multiple data sets into a single tidy data set.
This file will result in a tidy version of the “Human Activity Recognition Using Smartphones Dataset” supplied for the assignment.
library(plyr) ##need for join()
library(reshape2) ##need for melt() and dcast()
sessionInfo()## R version 3.0.3 (2014-03-06)
## Platform: x86_64-apple-darwin10.8.0 (64-bit)
##
## locale:
## [1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] reshape2_1.4 plyr_1.8.1
##
## loaded via a namespace (and not attached):
## [1] digest_0.6.4 evaluate_0.5.5 formatR_0.10 htmltools_0.2.4
## [5] knitr_1.6 Rcpp_0.11.1 rmarkdown_0.2.54 stringr_0.6.2
## [9] tools_3.0.3 yaml_2.1.11Data for this project was provided on the course website and can be downloaded from the following link: Data
After the data has been downloaded, place following files into working directory. They will be used to build a cohesive tidy data set:
This assignment required a single function, so I have commented the code of the single function instead of breaking the analyses down into a series of annotated steps. In summary, the function:
test and training data for subjectstest and training data for featuresWith all the data files in your working directory, run the run_analysis function to construct the tidy data set.
run_analysis<-function(){
#Read in subject IDs for train and test set
subject_test <- read.table("subject_test.txt", col.names=c("Subject"))
subject_train <- read.table("subject_train.txt", col.names=c("Subject"))
#Combine train and test sets into single data frame
subject_all <- rbind(subject_train, subject_test)
#Read in features data for train and test. Note: this is the data for each feature, not the feature labels.
features_test <- read.table("X_test.txt")
features_train <- read.table("X_train.txt")
#Combine training and test data for features into single data frame
features_data <- rbind(features_test, features_train)
#Read in list of features. Note: this is a list of the features names or labels. Not the features data
feature_list <- read.table("features.txt", col.names=c("index", "feature_labels"))
#Create 1 dimensional character vector containing feature labels from features_list data frame
feature_labels <- feature_list$feature_labels
##Create logical vector indicating columns which have mean() and std() in their name
features_subset <- grepl('mean\\(\\)|std\\(\\)',feature_labels)
#Create a character vector of only features with mean and standard deviation in their name
feature_list <- as.character(feature_labels[features_subset])
#Rename columns in features_data before subsetting out the mean and standard deviation columns, so that names match up still
colnames(features_data) <- feature_labels
## Extract only mean() and std() columns from features_data using the logical vector "features_subset"
features_data <- features_data[,features_subset]
#Read in activities for train and test
activities_test <- read.table("Y_test.txt")
activities_train <- read.table("Y_train.txt")
#Combine training and test activities into single data frame and rename column to "activity" from "V1"
activities_all <- rbind(activities_test, activities_train)
colnames(activities_all) <- "activityLabel"
## Recode activity values as descriptive names using the activity labels file
activity_labels<-read.table("activity_labels.txt",sep=" ",col.names=c("activityLabel","Activity"))
activities_all<-join(activities_all,activity_labels,by="activityLabel",type="left")
#Drop activity numbers
activities_all$activityLabel <- NULL
#Combine Actitivies, Subjects and Features all into one data frame
all_df <- cbind(features_data, activities_all, subject_all)
#Melt data frame for reshaping
tdf <- melt(all_df, id=c("Subject", "Activity"), measure.vars=feature_list)
#Reshape into tidy data frame by mean using the reshape2 package
tdf <- dcast(tdf, Activity + Subject ~ variable, mean)
#Reorder by Subject then Activity
tdf <- tdf[order(tdf$Subject, tdf$Activity),]
#Reindex Rows and move Subject to Column 1
rownames(tdf) <- 1:nrow(tdf)
tdf <- tdf[,c(2,1,3:68)]
#Output file
write.table(tdf,file="tidy_dataset.txt")
}