Intro
Data Science and AI might sound like the coolest thing that
there is right now. Whether you are a fresher trying to learn these skills or a
seasoned businessman who want to implement AI in your workflow, you might have heard
these terms thousands of times. But do you understand them properly? Do you
understand the differences? Because they are some of the most basic concepts
underlying. In this article, I’ll be hopefully able to answer the above
questions.
Definitions
Let me start with saying, I am not a big fan of definitions.
That is, I do not think anyone should ever try to memorize formal words
describing or defining something rather than actually understanding it. So
rather than giving formal definition, let’s see whether I can help you in
understanding the concepts.
Data Science literally means science that involves Data.
Anything that you do with Data, comes under the umbrella of ‘Data Science’. And
what is Data? Data is basically recorded information. We usually deal with
digital information, hence digital data. Anything that contains information is
data. So, the image you took a few days ago (which contains visual information)
is an example of Data. If you are reading this blog (which contains textual
information), this is also an example of Data. For certain reasons, we want to
access (or read) certain type of data, we might want to analyze them or want to
extract some insight or other information out of it. All these handling and
manipulation comes under Data Science, since all of these actions involve Data.
Now, AI means Artificial Intelligence. It is very hard to
define intelligence casually in a few words. But for the sake of simplicity, we
can equate intelligence with the ability to learn and apply your learning in
other situation. Though this is a crude way of defining intelligence, but if
you think thoroughly, this covers a large spectrum of intelligent behaviour.
When we discuss about intelligence, we usually refer to biological intelligence
which originated in nature through millions of years of evolution (a story for
later time!). But if you can mimic these same sort of behaviour artificially in
a machine (say, a computer or handheld device), then we call that Artificial
Intelligence.
Examples
Let’s give out a few hardcore examples before going to
differences.
A school collects different types of information regarding
its’ student population. It collects demographic information like name, age,
gender, address of residence, date of birth, guardian’s name and others. It
also collects academic information like all the subjects a student is taking,
their individual scores etc. . All these information can obviously be regarded
as Data. Now the school management might be interested in understanding more
about their students. So they might analyze this data and create visuals (like
charts, graphs etc.) based on these demographic and academic information they
have. The whole process may include- creating a database (a format to store the
data properly arranged), accessing the said database, data analysis and data
visualization. All of these tasks come under Data Science.
If you have a modern smartphone, it might have a ‘Face
recognition’ feature. That is, it can unlock the phone by recognizing your
face. If you step back and think a bit, we as humans also have this ability
mostly. Normally we can remember faces and then can recognize them later. This
is a sort of intelligent behaviour carried out by our brain. When we mimicked
the same ability in a machine (in this case, a mobile/ smartphone or laptop),
we call it Artificial Intelligence.
Difference
By now, you can understand that these two terms are not
entirely same. Or for some of you, it might seem totally different. But there
is a common link. If you are to build an AI system, that is, a machine with
intelligent behaviour, you do it using Data Science. You need data to teach a
machine a certain skill, thus making it intelligent. So AI invariably needs
Data Science. You can’t make an AI system without Data Science. But the
opposite is not true. Data science might cover a lot of other use cases which
do not need an intelligent behaviour, thus any AI. So,
All AI applications need Data Science but all Data Science use cases might not have AI.
Comments
Post a Comment