Weka Tutorial 17 Saving Results in Weka (Application) Video Lecture - Weka

hello and welcome to the waked tutorial
in this tutorial I'm going to show you
how you can save the results from wake
up into a comma separated value file or
CSV file in doing so we will take the
help of wake agree chooser so we click
on wake agree chooser and this time we
will will choose the experimenter
application so when we can experiment
environment is open at that time we are
with the radio button simple we have a
button named new and we click on Neal
so this stage we are dealing with the
datasets add new and we will take the
data file iris dot ARF that's provided
with Weka workbench directory and for
example we will try to use the algorithm
called naive Bayes we go to the
classifier naive Bayes we selected that
and click on OK in iteration control I'm
going to consider just one iteration and
we will experiment with the type
cross-validation with number of false
ten and of course our task is
classification not regression that's why
this classification radio button has
been chosen so when it is done we are
left with one single thing the
destination file so in Weka experiment
environment you can either save your
results in our file or in CSV file also
we have an option to save it in up
database for which you need to have a
JDBC connection with the database
so for the simplicity sake of the
tutorial I'm just showing them how you
can save the results into an ARFF file
or and
so into an csv file so for the first
time it will select CSV file and we are
naming the file experiment or you can
also select the desired place in your
hard disk where this experiment file is
going to be saved so we are on the
desktop experiment ok so click OK and
then we go to the run tab and we have a
start button here you click that so it
started and finished and there was no
error so when we are done with this you
can see that on my desktop we have a
comma separated value CSV file called
experiment so you can open that by
double clicking that and it will be open
with Microsoft Excel so here you can see
that we have numbers of columns and
those columns are also designated with
labels first one is data set the second
column named B is the number of run C is
the number of fold D is the scheme that
we used that the classifier we used as
an eye base and we have two other
informations here as well for example
percentage correct percentage incorrect
Kappa statistics absolute errors root
mean squared errors and so on so in this
way you can save the results provided by
way chi into a comma separated value
file so let's take a look how you can do
that to save the results into an art
file so on the setup tab of wake
experiment environment we choose the art
file as our file format and also we are
saving the same at the same place
desktop with the same name experiment
but only difference is the file is now
dot ARF
by keeping all other things as the same
as our previous set up with I distort
our file as our data set and nine base
arms our algorithms with 10-fold
cross-validation with one iteration we
then again go to the run tab and click
on start so as it as it finishes you can
see that on desktop you have an hour
file called experiment R if you click
right click on that and we edit that
with notepad plus plus then you can see
that this is the results what we've got
so far from wake up so this is the same
file that produced during our first run
with CSV the only difference is that it
has been made it has been converted into
an hour file and most importantly you
can use either of the file the Microsoft
Excel file dot CSV file or this dot R
file you can reuse them in your future
experiment hope that helps and I'll be
back with loads more with wake up thank
you