Data Analysis Project 2

Step 1: Data Collection

George Fisher george@georgefisher.com

Observations:

The data is drawn from the accelerometers of Samsung Galaxy S3 smartphones carried by 21 people numerically identified as subjects 1 3 5 6 7 8 11 14 15 16 17 19 21 22 23 25 26 27 28 29 30
Six activities are identified: laying, sitting, standing, walk, walkdown, walkup
There are no NAs in the initial dataset:
- 7352 obserations
- 563 variables
Per the prompt, training and test datasets are subsetted out and saved for later use
Training dataset:
- subjects 1, 3, 5, and 6
- 1315 observations
- 563 variables
Test dataset:
- subjects 27, 28, 29, and 30
- 1485 observations
- 563 variables

Download the dataset and look at it

load("samsungData.rda")
str(samsungData[c(1:20, length(samsungData) - 1, length(samsungData))])

## 'data.frame':    7352 obs. of  22 variables:
##  $ tBodyAcc-mean()-X  : num  0.289 0.278 0.28 0.279 0.277 ...
##  $ tBodyAcc-mean()-Y  : num  -0.0203 -0.0164 -0.0195 -0.0262 -0.0166 ...
##  $ tBodyAcc-mean()-Z  : num  -0.133 -0.124 -0.113 -0.123 -0.115 ...
##  $ tBodyAcc-std()-X   : num  -0.995 -0.998 -0.995 -0.996 -0.998 ...
##  $ tBodyAcc-std()-Y   : num  -0.983 -0.975 -0.967 -0.983 -0.981 ...
##  $ tBodyAcc-std()-Z   : num  -0.914 -0.96 -0.979 -0.991 -0.99 ...
##  $ tBodyAcc-mad()-X   : num  -0.995 -0.999 -0.997 -0.997 -0.998 ...
##  $ tBodyAcc-mad()-Y   : num  -0.983 -0.975 -0.964 -0.983 -0.98 ...
##  $ tBodyAcc-mad()-Z   : num  -0.924 -0.958 -0.977 -0.989 -0.99 ...
##  $ tBodyAcc-max()-X   : num  -0.935 -0.943 -0.939 -0.939 -0.942 ...
##  $ tBodyAcc-max()-Y   : num  -0.567 -0.558 -0.558 -0.576 -0.569 ...
##  $ tBodyAcc-max()-Z   : num  -0.744 -0.818 -0.818 -0.83 -0.825 ...
##  $ tBodyAcc-min()-X   : num  0.853 0.849 0.844 0.844 0.849 ...
##  $ tBodyAcc-min()-Y   : num  0.686 0.686 0.682 0.682 0.683 ...
##  $ tBodyAcc-min()-Z   : num  0.814 0.823 0.839 0.838 0.838 ...
##  $ tBodyAcc-sma()     : num  -0.966 -0.982 -0.983 -0.986 -0.993 ...
##  $ tBodyAcc-energy()-X: num  -1 -1 -1 -1 -1 ...
##  $ tBodyAcc-energy()-Y: num  -1 -1 -1 -1 -1 ...
##  $ tBodyAcc-energy()-Z: num  -0.995 -0.998 -0.999 -1 -1 ...
##  $ tBodyAcc-iqr()-X   : num  -0.994 -0.999 -0.997 -0.997 -0.998 ...
##  $ subject            : int  1 1 1 1 1 1 1 1 1 1 ...
##  $ activity           : chr  "standing" "standing" "standing" "standing" ...

# what activites? distributed how?
table(samsungData$activity)

## 
##   laying  sitting standing     walk walkdown   walkup 
##     1407     1286     1374     1226      986     1073

# how many subjects?
length(unique(samsungData$subject))

## [1] 21

# identified how?
unique(samsungData$subject)

##  [1]  1  3  5  6  7  8 11 14 15 16 17 19 21 22 23 25 26 27 28 29 30

# no incomplete cases
complete_cases = samsungData[complete.cases(samsungData), ]
dim(samsungData) == dim(complete_cases)

## [1] TRUE TRUE

Prompt

Your task is to build a function that predicts what activity a subject is performing based on the quantitative measurements from the Samsung phone.

For this analysis your training set must include the data from subjects 1, 3, 5, and 6. But you may use more subjects data to train if you wish.

Your test set is the data from subjects 27, 28, 29, and 30, but you may use more data to test. Be careful that your training/test sets do not overlap.

Create & save training and test datasets

train = subset(samsungData, samsungData$subject == 1 | samsungData$subject == 
    3 | samsungData$subject == 5 | samsungData$subject == 6)
table(train$subject)

## 
##   1   3   5   6 
## 347 341 302 325

dim(train)

## [1] 1315  563

numericActivity.train <- as.numeric(as.factor(train$activity))
table(train$activity)

## 
##   laying  sitting standing     walk walkdown   walkup 
##      221      198      227      266      193      210


test = subset(samsungData, samsungData$subject == 27 | samsungData$subject == 
    28 | samsungData$subject == 29 | samsungData$subject == 30)
table(test$subject)

## 
##  27  28  29  30 
## 376 382 344 383

dim(test)

## [1] 1485  563

numericActivity.test <- as.numeric(as.factor(test$activity))
table(test$activity)

## 
##   laying  sitting standing     walk walkdown   walkup 
##      293      264      283      229      200      216


save(train, test, numericActivity.train, numericActivity.test, file = "train_and_test.Rda")

Info about the system running this code

print(str(.Platform))

## List of 8
##  $ OS.type   : chr "windows"
##  $ file.sep  : chr "/"
##  $ dynlib.ext: chr ".dll"
##  $ GUI       : chr "RTerm"
##  $ endian    : chr "little"
##  $ pkgType   : chr "win.binary"
##  $ path.sep  : chr ";"
##  $ r_arch    : chr "x64"
## NULL

print(version)

##                _                           
## platform       x86_64-w64-mingw32          
## arch           x86_64                      
## os             mingw32                     
## system         x86_64, mingw32             
## status                                     
## major          3                           
## minor          0.2                         
## year           2013                        
## month          09                          
## day            25                          
## svn rev        63987                       
## language       R                           
## version.string R version 3.0.2 (2013-09-25)
## nickname       Frisbee Sailing

print(sessionInfo(), locale = FALSE)

## R version 3.0.2 (2013-09-25)
## Platform: x86_64-w64-mingw32/x64 (64-bit)
## 
## attached base packages:
## [1] stats     graphics  grDevices utils     datasets  methods   base     
## 
## other attached packages:
## [1] knitr_1.5
## 
## loaded via a namespace (and not attached):
## [1] evaluate_0.5.1 formatR_0.10   stringr_0.6.2  tools_3.0.2

print(Sys.time())

## [1] "2013-11-27 18:14:12 EST"