A series of experiments have been carried out with a group of 30 volunteers within an age bracket of 19-48 years. Each person performed six activities (WALKING, WALKING_UPSTAIRS, WALKING_DOWNSTAIRS, SITTING, STANDING, LAYING) wearing a smartphone (Samsung Galaxy S II) on the waist. Using its embedded accelerometer and gyroscope, we captured 3-axial linear acceleration and 3-axial angular velocity at a constant rate of 50Hz. The experiments have been video-recorded to label the data manually. The obtained dataset has been randomly partitioned into two sets, where 70% of the volunteers was selected for generating the training data and 30% the test data.
The sensor signals (accelerometer and gyroscope) were pre-processed by applying noise filters and then sampled in fixed-width sliding windows of 2.56 sec and 50% overlap (128 readings/window). The sensor acceleration signal, which has gravitational and body motion components, was separated using a Butterworth low-pass filter into body acceleration and gravity. The gravitational force is assumed to have only low frequency components, therefore a filter with 0.3 Hz cutoff frequency was used. From each window, a vector of features was obtained by calculating variables from the time and frequency domain. See features_info.txt for more details.
For each record it is provided:
- Triaxial acceleration from the accelerometer (total acceleration) and the estimated body acceleration.
- Triaxial Angular velocity from the gyroscope.
- A 561-feature vector with time and frequency domain variables.
- Its activity label.
- An identifier of the subject who carried out the experiment.
The dataset includes the following files:
-
README.txt: This file.
-
features_info.txt: Shows information about the variables used on the feature vector.
-
features.txt: List of all features.
-
activity_labels.txt: Links the class labels with their activity name.
-
train/X_train.txt: Training set.
-
train/Y_train.txt: Training labels.
-
test/X_test.txt: Test set.
-
test/Y_test.txt: Test labels.
The following files are available for the train and test data. Their descriptions are equivalent.
-
train/subject_train.txt: Each row identifies the subject who performed the activity for each window sample. Its range is from 1 to 30.
-
train/Inertial_Signals/total_acc_x_train.txt: The acceleration signal from the smartphone accelerometer X axis in standard gravity units 'g'. Every row shows a 128 element vector. The same description applies for the 'total_acc_x_train.txt' and 'total_acc_z_train.txt' files for the Y and Z axis.
-
train/Inertial Signals/body_acc_x_train.txt: The body acceleration signal obtained by subtracting the gravity from the total acceleration.
-
train/Inertial Signals/body_gyro_x_train.txt: The angular velocity vector measured by the gyroscope for each window sample. The units are radians/second.
This study has been reported in [1].
[1] Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra and Jorge L. Reyes-Ortiz. Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine. International Workshop of Ambient Assisted Living (IWAAL 2012). Vitoria-Gasteiz, Spain. Dec 2012
The program should do basically the following:
- Merge the training and the test sets to create one data set.
- Extract only the measurements on the mean and standard deviation for each measurement.
- Use descriptive activity names to name the activities in the data set
- Appropriately label the data set with descriptive variable names.
- From the data set in step 4, create a second, independent tidy data set with the average of each variable for each activity and each subject.
The program basically creates two tables, although it also creates other ancillary
tables in the process. The two final tables created by the program have been named
data_tbl and smy_tbl. The first one contains all information from the training
and test
data sets, while the second one contains a summary of all "mean" and "standard deviation"
features averaged by subject and activity.
The following section explains the program code as it is in the run_analysis.R script,
except for the dim() instructions.
Load all R packages required
library(data.table)Read all data sets from corresponding data files into appropriate data tables. Data sets are stored in files kept in a folder structure as follows:
<working folder>--+
+-- features.txt
+-- train +
+--X_train.txt
+--Y_train.txt
+--subject_train.txt
+-- test +
+--X_test.txt
+--Y_test.txt
+--subject_test.txt
Load variables' names table (features), used in both training
and test data sets. This feature table has two columns: <feature no.>, and
<feature name>.
f_tbl <- read.table("features.txt")Load training related data sets. Table names are self-explanatory.
x_trn <- read.table("./train/X_train.txt")
y_trn <- read.table("./train/Y_train.txt")
s_trn <- read.table("./train/subject_train.txt")Label training data sets variables with significative names.
colnames(s_trn)[1] <- "Subject_ID"
colnames(y_trn)[1] <- "Activity_ID"
names(x_trn) <- t(f_tbl[2]) # transpose column vector of names to make it a rowLoad test related data sets. Table names are self-explanatory.
x_tst <- read.table("./test/X_test.txt")
y_tst <- read.table("./test/Y_test.txt")
s_tst <- read.table("./test/subject_test.txt")Label test data sets variables with significative names.
colnames(s_tst)[1] <- "Subject_ID"
colnames(y_tst)[1] <- "Activity_ID"
names(x_tst) <- t(f_tbl[2]) # transpose column vector of names to make it a rowAppend row-wise, corresponding x, y and subject data sets
from training and test data sets. Table names are self-explanatory.
xs_tbl <- rbind(x_tst,x_trn)
ys_tbl <- rbind(y_tst,y_trn)
ss_tbl <- rbind(s_tst,s_trn)Merge by columns the three previous tables so as to get one big data table with all the information from all subjects, activities and features.
data_tbl <- cbind(ss_tbl,ys_tbl,xs_tbl)Load activity table, which has columns <activity No.> and <activity name>,
and change data_tbl activity coding numbers to significative names.
data_tbl is the first data table looked for.
act_tbl <- read.table("activity_labels.txt") # columns of table are <number> and <name>
for(i in 1:length(data_tbl[,2])) { # column 2 of 'data_tbl' is "Activity_ID"
data_tbl[i,2] <- as.character(act_tbl[data_tbl[i,2],2]) # change activity number to activity name
}Let's check the dimensions out of this table, just to verify we have all valid data.
Note that the number of observations corresponds to the sum of the number of
observations in the trainig data set + the number of observations in the test
data set. Also, the number of columns corresponds to the sum of the number
of features + one column for the subjects and one column for the activities.
dim(data_tbl)Get a summarized tidy data set containing the average of all and only
-mean() and -std() features for each subject and each activity.
Extract from total data table data_tbl all variables corresponding to all and only
mean and standard deviation features, for all subjects and activities. We name
this table sub_tbl.
sub_tbl <- cbind(data_tbl[,1:2], # columns 1 and 2 of 'data_tbl' are "Subject_ID" and "Activity_ID"
subset(data_tbl,select=grep("-mean()",names(data_tbl), # "mean" features
fixed=TRUE)),
subset(data_tbl,select=grep("-std()",names(data_tbl), # "standard deviation" features
fixed=TRUE))
)Create an empty data table containing only the variable names of all
selected features in the sub_tbl created in the previous step,
to enable gradual appending of value rows in the summary table smy_tbl
being constructed.
Note: smy_tbl stands for "summary table".
smy_tbl <- as.data.table(setNames(replicate(length(names(sub_tbl)),
numeric(0),
simplify = FALSE),
names(sub_tbl)
)
)Final step. Fill in the summary table smy_tbl as follows:
For each `subject` and each `activity`:
extract the corresponding subtable,
create an auxiliary table with the average of each feature, and
append this auxiliary table to the end of the summary table
Note: The follwing table shows the meaning of each variable used in the the next code fragment.
| name | meaning |
|---|---|
| sbj | "subject" |
| act | "activity" |
| ssa | "subset subject-activity" |
| som | "summary of means (averages)" |
for (sbj in 1:30){
for (act in act_tbl[,2]){ # column 2 of 'act_tbl' has the activity names
# subset table by 'subject' and 'activity'
ssa <- subset(sub_tbl,Subject_ID==sbj & Activity_ID==act)
# create auxiliary table with the average value of each feature
som <- data.table(Subject_ID=sbj, Activity_ID=act,
matrix(apply(ssa[-c(1:2)], 2, mean),
ncol=length(names(sub_tbl))-2
)
)
# set significant variable names to 'som' table just created
setnames(som,names(sub_tbl))
# append 'som' table to the end of the summary table 'smy_tbl'
smy_tbl <- rbind(smy_tbl,som,use.names=TRUE)
}
}smy_tbl is the final data table looked for.
Finally, we could check the dimensions out of this table, just to verify we have all valid data.
Note that the number of rows corresponds to number of subjects times the number of
activities, i.e. 30*6=180, while the numbers of columns corresponds to the sum
of the number of features having -mean() and -std() in their names.
dim(smy_tbl)