How to Build a Data Mining Model from Scratch (Part 1 of 3)

How to Build a Data Mining Model from Scratch (Part 1 of 3)


Welcome back to the Data Professor, I’m
Chanin Nantasenamat and in this episode I’m going to give you a quick
introduction about what is data science and how you can go about building your
very first prediction model so without further due, let’s get started! Data is ubiquitous, and in this day and
age, we have an ever-increasing amount of data, infamously known as big data, which we can use to analyze, to gain insights and to drive the decision-making process. So, what exactly is data, data pertains to
information about entities of interest For example, (1) health parameters of a human being such as the red and white blood cell count, the blood profile, lipid
profile and other parameters that describes the health status of an
individual, (2)characteristics of cars such as the top speed that it can go and fuel
consumption rates, (3) properties of drugs such as the molecular size, solubility,
electronic and hydrophobic properties of the drug. Simply put, data science is a very big
field that encompasses several smaller disciplines such as statistics,
mathematics, data visualization, programming, data mining and machine learning. So as you can see data mining is a subset of data science and it refers to
the specific process of making use of the data in order to build a prediction
model and extracting knowledge from the data, while machine learning refers to
the learning algorithms that are used to create the prediction models inside the
data mining process. So there you have it a very brief introduction to data
science. Now comes the fun part, let’s get started
in building our very first prediction model! WEKA is a program for performing data
mining. It has an intuitive graphical user interface that allows you to
pre-process, transform the data as well as construct the prediction model using
a variety of machine learning algorithms and it was created by two developers Ian
Witten and Eibe Frank from the University of Waikato. So let’s begin by first
installing WEKA onto your computer. So what you need to do is go to
Google and then search for WEKA and then click on the first link. So notice
that the URL is coming from the University of Waikato. So click on the
link. So it’s the page that was open a couple of seconds ago. So let’s get
started by downloading the program. So click on the download button and then
scroll down, you’ll notice that they’re going to have several versions
here. Snapshot is when they have a, like a beta version, which is not stable yet, but
what you want is the stable version right here or they also have the
developer version where they also provide new features, which are not yet
stable but are included for your usage here. If you’re into the latest feature
you might want to try this one. But if you’re starting out, I would recommend
using the stable version. So it has (support) for many platform: on the Windows platform
for the Mac platform and also for Linux platform as well. So before you begin you
will have to select one of the four links right here. So what are they? Well,
the first link is the WEKA program, right here, version 3.8.3 and it also comes with a Java Virtual Environment as you can see from
the final name, for the 64-bit version. However, the second file is the WEKA
program alone as you can see here by the name of WEKA and the version number
3.8.3 and then x64 would mean it is built for the 64-bit version
of Windows but it does not come with the Java Virtual Machine so therefore you
don’t see the JRE in the file name. And the third file
is similar to the first file in which it has the WEKA program along with the
Java Virtual Machine but it is built for the 32-bit version of your Windows. And
the fourth file is the WEKA program built for the 32-bit version. So if you
are wondering which version should you go with? Well, let’s check out what is the
version of your computer’s (Java)? whether it is 64 or 32 bit, Oh it’s right here Properties and then notice the 64-bit
version right here. So this computer has 64-bit, so I’m going to go for
the 64-bit version, however I will have to identify whether I want to have Java
or without Java. So in order to do that let’s check whether my computer has Java
or not and you can do the same by going to the search icon, type in CMD and click
on the command prompt and then you will see this command prompt
window coming up, type in Java and if it says that Java is not recognized, then it
means that your computer does not have Java installed. So let’s go with the
first file which has Java prepackaged along with the WEKA software. So let’s
click on here and that will take you to the download link. Wait a bit, okay and then your download have started so it’s a 115 megabytes so that
should take you a little while, okay so the internet speed is going up and we are a
couple of seconds away from downloading the program. Okay so it’s finished and
let’s install. So click on the installation file and it will ask
whether you want to allow this program to make changes to your device. So I’ll
click on Yes and then the next step is pretty easy and straightforward. So click
on the Next button and we are close to completion and now
it’s going to install the Java Virtual Machine. So click on the install
button, click on okay, wait some more. Okay so we’re almost there, okay so Java has
successfully been installed and I will click on the Close button and then WEKA
say it is completed, so once it’s completed, we’ll click on the Next
button and then it has to tick for us to start WEKA and click on Finished.
So, until next time, I’m Chanin Nantasenamat on the Data Professor channel and
if you haven’t subscribed yet, please consider subscribing and clicking on the
notification bell so that you will be notified on the next video. So, I’ll see
you in the next one!

1 thought on “How to Build a Data Mining Model from Scratch (Part 1 of 3)”

Leave a Reply

Your email address will not be published. Required fields are marked *