THE FUTURE STARTS HERE

Different types of Data

Analytics

HARNESS THE POWER OF ANALYTICS

CONVERT RAW DATA INTO ACTIONABLE INSIGHTS

SUPPORTED BY
WHY LEARN DATA ANALYTICS?

• As more and more companies move to the cloud, the need for data science
and engineering talent will further increase

• Rapid growth and demand in digital transformation and advanced

analytics have led to a surge in demand for data science professionals in
key technology areas such as artificial intelligence

• Analytics enables data-driven decision-making which is vital for

corporations today.
 UBER
Skype Passengers take
Users Make 694 Rides Facebook
Users Like
110,040 Calls
4,166,667 Posts
Buzzfeed Twitter
Users View Users Send
34,150 Videos 347,222 Tweets
Snapchat
HOW MUCH Users Share
YouTube
Users Upload
284,722 Snaps
DATA IS GENERATED? Every 300 Hours of
new video
Data is being created all the time without us even Tinder Minute of
Users Swipe Instagram
noticing it. Much of what we do every day now happens 590,278 Times the Day Users Like
in the digital realm, leaving an ever-increasing digital 1,736,111 Photos
trail that can be measured and analyzed. Vine
Users Play Pinterest
1,041,666 Videos Users Pin
Amazon 9,722 Images
Receives
Apple
4,310 Unique Users Download
Visitors Reddit Netflix 51,000 Apps
Users CAst
Subscribers Stream
18,327 Votes
77,160 Hours of
Video
DATA SCIENCE IN VARIOUS INDUSTRIES

Retail Healthcare
− Analyze Consumer Behaviors − Detect Causes of disease
− Predict Product Pricing − Risk prediction

Medical Enterprise
− Disease prediction − Capturing data volume, variety
− Analyze patient recovery growth and value

Insurance Machine Learning

− Fraud and risk detection − Data science is leveraged to
− Insurance claims prediction disrupt ML based application

Funding Digital Marketing

− Detect possible fundraisers − Detect targeted audience
− Demographic behavior analysis
The Four Types :
Descriptive Analytics

The scope of Descriptive Analytics is to describe what has

happened and primarily looks at past data to understand
patterns and trends in the data analysed.

It serves as a base point or starting point for an analytics

strategy.

It utilizes techniques such as data mining, summary

statistics, and clustering to provide past trends.
The Four Types :
Diagnostic Analytics

Why did this thing happen is the question that falls under
the arena of Diagnostic Analytics.

The basic purpose of this analytics is to find out why a

thing has happened or lay the foundation for the reasons
supporting that event. In this type of analysis, we basically
consider the following:

• Why the event has happened

• Rooted in the past and is based on the concept of
probability

This type of analytics involves comparing all the coexisting

trends and the movements by uncovering the correlation
between certain variables. Moreover, it also determines
casual relationships wherever it is possible.
This Photo by Unknown Author is licensed under CC BY-SA-NC
The Four Types :
Predictive Analytics

This analysis basically looks at the question of What may

happen in future? This type of analytics is basically
centred around:

• What can happen under a certain condition

• This analytics looks into the future and is also based on
probability

It analyses all the historical trends in tandem with the

industry trends to make informed predictions regarding
This Photo by Unknown Author is licensed under CC BY-SA
the future. Here, in this analytics, historical data and other
variables like economic forecast, industry trends and
consumer confidence.

This Photo by Unknown Author is licensed under CC BY-SA-NC

The Four Types :
Prescriptive Analytics

The prescriptive analytics basically revolved around the

question of ‘what would be your next move’ or ‘how to do a
thing’?

It takes into account all the needed variables that you can
logically anticipate. The agenda of this analytics is:

Based on the required outcomes, what type of actions are

required
This analytics looks into the future and is anchored
towards a certain set of rules
It uses the techniques of simulation analysis, artificial
intelligence and neural networks

This analytics is built on the base of predictive analytics

that determines prescribed actions on the basis of desired
outcomes to help achieve business objectives.
 PROBLEM EXAMPLE (USING PYTHON)
Written Task
Write a one-page memo describing your
color and format choices and how you
customized this visualization to fit the
needs of your audience

Objectives
§ To see the status of the project
§ Attempting to see the performance of
the country
§ Attempting to see the performance of
the submitting officer

Color Selection
To use color to determine the size of
project

Format Selection
Click on the size of the project will be able to
see the submitting officer and the respective
country involved and size of the total project.
 Top 5 Countries w Highest numbers of Top
100 Units
The. Majority of Top 100 Universities come from USA, who
has about 8x more top 100 universities than the next best
country: the UK. The difference between the number of top
100 universities between the UK, Japan, France, and
Switzerland is less drastic compared to the disparity
between USA and UK.

With only integers, it's difficult to see if USA is the best

place to go for a quality education because we should be
using a percentage figure of: (Number of world rank Top
100 schools / Total Universities in country) *100 to get a
more accurate representation.
THE FUTURE STARTS HERE

The Four Types of Data

Analytics
THANK YOU
Thank You for taking the First Step.
Convert raw data into actionable information

SUPPORTED BY