Data & Analytics Exam > Data & Analytics Videos > Weka Tutorial > Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation)
Weka Tutorial 27 Inverse k-fold Cross Validation (Model Evaluation) Video
Text Transcript from Video
so hello everyone welcome to another
tutorial on wake up in this tutorial I'm
going to show you a very very very
interesting stuff while you wake up the
in regular releases by University of
Waikato the regular version of wake up
it has an option called cross-validation
with number of foals to be chosen for
example one one cannot be chosen there
can be two can be three can be for up to
ten or twenty nine whatever you say
anything greater than one will be okay
there what does it mean in order to do
that I need to go to the Explorer so on
explorer I'm going to open a file any
random or file for this tutorial so
that's my random or file it has been
chosen
I go to classify tab I select any
algorithm for example nine base and you
can see that the ticking test options
you have a cross-validation sub-option
and you have something to put here as
false so ten is default there but you
can put one and click start to get this
error message called problem evaluating
classifier number of holes must be
greater than one what does it mean well
if you have your fold equal to 1 then it
means that you are actually trying to
really evaluate the entire data set
there is no concept of k-fold
cross-validation here okay so we click
OK and we select two what it means is
wicca is going to partition your data
set into two parts one part will be used
as training data and the other part will
be used as test data in other words you
are having 50% of your data as your
training data and 50% of your data as
your test data so what happens if we put
3 there this means that your data set
will be partitioned into three equal
size partitions and it also means that
you are having 66 percent of your data
as your training data where 33% of the
data as your test
finally with Jen falls this means that
you are utilizing 90% of your data in
your data set as your training data and
10% of them will be used as test data
but what happens when the situation is
like this that you are trying to have
10% of your data as your training data
and 90% of your data as your test data
that is not possible with Waker with at
least this release of Weka from
University of Waikato they don't have
this concept of inverse k-fold
cross-validation okay so what do we do
how are you going to achieve an inverse
k-fold cross-validation where I have
extracted the jar files and you know
that I have to click on the collective
classification jar so as I open that
this version of Weka
then after opening that I'm clicking on
the Explorer button and I'm opening the
same file so I go back to my drive where
I'm having my our files for this
tutorial select it right now I go back
to classify tab on choosing a random
classifier the best one so far is naive
Bayes because it's easy to find and you
can see that we are having in the test
options a sub option called
cross-validation and it's set to default
10 so if we want to do an inverse
10-fold cross-validation we just need to
put an AI and AI before 10 that's it
what it means is waker is partitioning
your data set into 10 equal sized folds
and 10% of those data will be used as
training data but 90% of the data will
be used as as your test data so that's
completely opposite of this okay so now
to demonstrate I'm having 10-fold
cross-validation I'm starting the
classic
education tasks and I see that well the
if measure is 79.5%
okay now I'm changing the situation here
I'm conversing my setup here by putting
an eye
this means that previously you had 90
percent of my data as your trip asthma
as training data and ten percent of the
data as test data now it's going to be
the opposite
ten percent training ninety percent test
and I click on start so when it is done
you can see that we are having a
different set of results where my f
measure is seventy nine point six where
previously it was seventy nine point
five they are close but still there are
some differences in the values of
precision recall and F measure and so on
so have fun guys because this concept is
very important because it's not always
the case
you're going to have a 10-fold
cross-validation sometimes it is
required to use inverse 10-fold
cross-validation in general inverse
k-fold cross-validation I hope that
helps and thank you for watching I'll be
right back with more tutorials thank you
you
tutorial on wake up in this tutorial I'm
going to show you a very very very
interesting stuff while you wake up the
in regular releases by University of
Waikato the regular version of wake up
it has an option called cross-validation
with number of foals to be chosen for
example one one cannot be chosen there
can be two can be three can be for up to
ten or twenty nine whatever you say
anything greater than one will be okay
there what does it mean in order to do
that I need to go to the Explorer so on
explorer I'm going to open a file any
random or file for this tutorial so
that's my random or file it has been
chosen
I go to classify tab I select any
algorithm for example nine base and you
can see that the ticking test options
you have a cross-validation sub-option
and you have something to put here as
false so ten is default there but you
can put one and click start to get this
error message called problem evaluating
classifier number of holes must be
greater than one what does it mean well
if you have your fold equal to 1 then it
means that you are actually trying to
really evaluate the entire data set
there is no concept of k-fold
cross-validation here okay so we click
OK and we select two what it means is
wicca is going to partition your data
set into two parts one part will be used
as training data and the other part will
be used as test data in other words you
are having 50% of your data as your
training data and 50% of your data as
your test data so what happens if we put
3 there this means that your data set
will be partitioned into three equal
size partitions and it also means that
you are having 66 percent of your data
as your training data where 33% of the
data as your test
finally with Jen falls this means that
you are utilizing 90% of your data in
your data set as your training data and
10% of them will be used as test data
but what happens when the situation is
like this that you are trying to have
10% of your data as your training data
and 90% of your data as your test data
that is not possible with Waker with at
least this release of Weka from
University of Waikato they don't have
this concept of inverse k-fold
cross-validation okay so what do we do
how are you going to achieve an inverse
k-fold cross-validation where I have
extracted the jar files and you know
that I have to click on the collective
classification jar so as I open that
this version of Weka
then after opening that I'm clicking on
the Explorer button and I'm opening the
same file so I go back to my drive where
I'm having my our files for this
tutorial select it right now I go back
to classify tab on choosing a random
classifier the best one so far is naive
Bayes because it's easy to find and you
can see that we are having in the test
options a sub option called
cross-validation and it's set to default
10 so if we want to do an inverse
10-fold cross-validation we just need to
put an AI and AI before 10 that's it
what it means is waker is partitioning
your data set into 10 equal sized folds
and 10% of those data will be used as
training data but 90% of the data will
be used as as your test data so that's
completely opposite of this okay so now
to demonstrate I'm having 10-fold
cross-validation I'm starting the
classic
education tasks and I see that well the
if measure is 79.5%
okay now I'm changing the situation here
I'm conversing my setup here by putting
an eye
this means that previously you had 90
percent of my data as your trip asthma
as training data and ten percent of the
data as test data now it's going to be
the opposite
ten percent training ninety percent test
and I click on start so when it is done
you can see that we are having a
different set of results where my f
measure is seventy nine point six where
previously it was seventy nine point
five they are close but still there are
some differences in the values of
precision recall and F measure and so on
so have fun guys because this concept is
very important because it's not always
the case
you're going to have a 10-fold
cross-validation sometimes it is
required to use inverse 10-fold
cross-validation in general inverse
k-fold cross-validation I hope that
helps and thank you for watching I'll be
right back with more tutorials thank you
you
About this Video
4.89/5 Rating
Apr 29, 2026 Last updated
Related Exams
Video Description: Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) for Data & Analytics 2026 is part of Weka Tutorial preparation. The notes and questions for Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) have been prepared according to the Data & Analytics exam syllabus. Information about Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) covers all important topics for Data & Analytics 2026 Exam. Find important definitions, questions, notes, meanings, examples, exercises and tests below for Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation).
Introduction of Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) in English is available as part of our Weka Tutorial for Data & Analytics & Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) in Hindi for Weka Tutorial course. Download more important topics related with notes, lectures and mock test series for Data & Analytics Exam by signing up for free.
Description
Video: Weka Tutorial 27: Inverse k of Weka Tutorial have been curated by the Data & Analytics experts, helping you revise the topic quickly for exam preparation. Watch on EduRev.
Information about Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation)
Here you can find the meaning of Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) defined & explained in the simplest way possible. Besides explaining types of Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) theory, EduRev gives you an ample number of questions to practice Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation) tests, examples and also practice Data & Analytics tests.
Related Searches
ppt, MCQs, Free, study material, pdf , practice quizzes, Semester Notes, Sample Paper, Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation), Exam, shortcuts and tricks, Previous Year Questions with Solutions, Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation), Summary, Objective type Questions, mock tests for examination, video lectures, Important questions, Weka Tutorial 27: Inverse k-fold Cross Validation (Model Evaluation), past year papers, Extra Questions, Viva Questions;