Data Analytics

Tilani Gunawardena
PhD(UNIBAS), BSc.Eng(Pera), FHEA(UK), AMIE(SL)
2019/11/22
Why?

2
 Data
Data- Data is a set of values of qualitative or quantitative
variables. It is information in raw or unorganized form. It may
be a fact, figure, characters, symbols etc.
Information- Meaningful or organized data is information,
comes from analyzing data

3
Data Analytics
Analytics- Analytics is the discovery , interpretation, and communication
of meaningful patterns or summery in data.
Data Analytics (DA) is the process of examining data sets in order to draw
conclusion about the information it contains.
Analytics is not a tool or technology, rather it is the way of thinking and
acting on data.

Data on its own is useless unless you can make

sense of it!

4
Primary Focus Areas for Analytics

5
 Data Analytics - Example
5

 Business analytics
 Risk
 Fraud
 Health
 Web

6
Business Analytics

7
The Process of Statistical Analysis
When we have resource constraints, Statistical Analysis enables us to make quantitative
inferences based on an amount of information we can analyze (a sample).
Form Identify Data Prove/Disprove
Hypotheses Source Hypothesis

8
Data Analytics Life Cycle

9
 Types of Data Analytics
Descriptive Analytics
Predictive Analytics
Prescriptive Analytics

10
Types of DataAnalysis

11
 Types of DataAnalysis

Descriptive Predictive Prescriptive

• Aims to help • Helps forecast • Suggests conclusions or

uncover valuable behavior of people actions that may be
insight from the and markets taken based on the
data being analyzed • Answers the question analysis
• Answers the “What could happen?” • Answers the
question question
“What has • “What should be
happened?” done?”
• What Should we do

12
Types of Analytics

How does Netflix

frequently recommend
just the right movie?
How do grocery cashiers
Descriptive
know to hand you coupons
Analytics you might actually use?
insight into the past
Predictive Why do airline prices
Analytics change every hour?
understanding the future

Prescriptive
Analytics
advice on possible outcomes

13
 Descriptive DataAnalytics
 Though the most simple type, it is used most Mean, Median and Mode Amounts
often.
of Items Purchased
 Two types of descriptiveanalysis:
1. Measures of central tendency (tells us about 7 6.5
the middle) 6
 Mean − theaverage 5
 Median− the midpoint of the 4
responses 3
2
 Mode − the response with the highest 2
1
frequency 1
2. Measures of dispersion 0
Mean Median Mode
 Range −the min, the max and the
 distance between the two
 Variance − the average degree towhich each Customer_ID Items Purchased Amount Spent
of the points differ fromthe mean 29304 1 1.09
 Standard Deviation − the most 28308 3 44.43
common/standard way of expressing the 19962 21 218.58
spread of data 30281 1 73.02

14
Variability
No Variability in Cash Flow
Mean
Mean

Variability in Cash Flow Mean

Mean

15
16
 Predictive DataAnalytics

 Some mistake predictive analysis to have exclusive relevance to predicting

future events.
− However, in cases such as sentiment analysis, existing data (e.g., the text
of a tweet) is used to predict non-existent data (whether the tweet is positive
or negative).
 Several of the models that can be used for predictive analysis are:
− Forecasting
− Simulation
− Regression
− Classification
− Clustering

17
 Prescriptive DataAnalytics

 Decisions can be formulated from descriptive and predictive analysis

− If I need to cut a product and I know that product C is least preferred and least
profitable, I will cut productC.
 However, prescriptive analytics explicitly tell youthe decisions that should
be made. This can be done using a variety of techniques:
− Linear programming
− Integer programming
− Mixed integer programming
− Nonlinear programming

18
 ComparingtheThreeTypesofDataAnalytics
 Descriptive analysis is mostcommon.
− Best practice to perform descriptive
analyses prior toprescriptive/predictive
 Understand that distribution, variance,
skew, etc., may exclude certainmodels
 How to know which type of analysis to
pursue:
− How much time do you have?
− What resources are available toyou?
− How accurate is your data? How accurate
do you need the model/analysis to be?
− How popular/accepted is the model you are considering?
 Don’t subscribe to “that’s how we’ve always done it,” but
remember to use a model that stakeholders will accept.

19
Why Big Data Analytics?
Why is Big Data Analytics important?
Big data analytics helps organizations harness
their data and use it to identify new
opportunities. That, in turn, leads to smarter
business moves, more efficient operations,
higher profits and happier customers.

20
Data Analytics Tools
Enter Data Scientists
Data Scientist: A Business analyst is not able
to discover insights from huge
The sets of data of different
domains.

SEXIEST Data scientists can work in co-

Job
ordination with different
verticals of an organization
and find useful
patterns/insights for a

In The 21
company to make tangible
ST business decisions.

century
Harvard Business Review, Oct 2012
15,000%
INCREASE IN JOB POSTINGS FOR
DATA SCIENTISTS IN THE US
BETWEEN 2011-12

23
USE CASES
 Facebook
 2.5 billion monthly active users
 Deep learning: Facebook makes use of facial recognition and text
analysis
 Facial recognition, Facebook uses powerful neural networks to classify
faces in the photographs
 Uses its own text understanding engine called “DeepText” to understand
user sentences.
 For targeted advertising: Deep Learning
It uses the insights gained from the data to cluster users based on their
preferences and provides them with the advertisements that appeal to
them.

25
 Amazon
Amazon heavily relies on predictive analytics to increase customer satisfaction.
Personalized recommendation system.
This also comes through the suggestions that are drawn from the other users who use similar products or provide
similar ratings.
This also comes through the suggestions that are drawn from the other users who use similar products or provide
similar ratings.

Amazon’s Recommendation Engine

26
 Amazon
Fraud Detection
Amazon has its own novel ways and algorithms to
detect fraud sellers and fraudulent purchases.

27
 Uber
Uber is a popular smartphone application that allows you to book
a cab.
Uber makes extensive use of Big Data.
Uber has to maintain a large database of drivers, customers, and
several other records.
Whenever you hail for a cab, Uber matches your profile with the
most suitable driver.

28
Bank of America - Using Data to Leverage Customer Experience

BoA are making use of Data Science and predictive analytics

Banking industries are able to detect frauds in payments and customer
information
Prevents frauds regarding insurances, credit cards, and accounting.
Banks employ data scientists to use their quantitative knowledge where
they apply algorithms like association, clustering, forecasting, and
classification.
Risk modeling - Using Machine Learning, banks are able to minimize risk
modeling.

29
Bank of America - Using Data to Leverage Customer Experience

Understanding their customers through an intelligent customer segmentation

approach: high-value and low-value segments.
Clustering, logistic regression, decision trees to help the banks to understand
the Customer Lifetime Value (CLV) and take group them in the appropriate
segments.

30
 Airbnb

host accommodations as well as find them through its mobile app and website
Contain massive big data of customer and host information, homestays and
lodge records, as well as website traffic
Ex : In 2014, Airbnb found out that users from certain countries would click the
neighborhood link, browse the page and photos and not make any booking.
In order to mitigate this issue, Airbnb released a different version for the users
from those countries and replaced neighborhood links with the top travel
destinations. This saw a 10% improvement in the lift rate for those users.

31
 Spotify

online music streaming giant

With over 100 million users, Spotify deals with a massive amount of big data
uses the 600 GBs of daily data generated by the users to build its algorithms
to boost user experience.
In the year 2017, Spotify used data science to gain insights about which
universities had the highest percentage of party playlists and which ones spent
the most time on it.

32