Data Analytics

Data Analytics
Dr. Vala Ali Rohani
Vala@um.edu.my
VRohani@gmail.com

My Bio Data
• Postdoctoral scholar in Social Network Analysis
• PhD in Software Engineering (Recommender Systems)
• Social Network Analysis from University of Michigan
Professional Certificates:
• University Lecturer for more than 10 years
• Mining Massive Datasets from Stanford University
• Pattern Discovery in Data Mining from Illinois University
• Process Mining from Eindhoven University of Technology
• Statistical Analysis using SPSS & SAS from University of Malaya
• MongoDB for DBAs from MongoDB
Data Analytics by Vala Ali Rohani

Presentation outline
Data Science & Big Data
Social Network Analysis
Process Mining
Market Basket Analysis

Data Analytics
&
Big Data

Domain Terminology
• Data Analysis, Data Mining, Machine Learning and Mathematical Modeling are
tools: means towards an end.
• Analytics, Business Intelligence, Econometrics and Artificial Intelligence are
application areas: domains that use the tools above (and others) to produce results
within its subject.
• Statistics is a branch of Mathematics providing theoretical and practical support to the
above tools.
• Data Science is a catch-all term to describe using those all tools to provide answers in
those all areas (and also in others), specially when dealing with Big Data
http://www.quora.com/What-is-the-difference-between-Data-Analytics-Data-Analysis-Data-Mining-Data-Science-Machine-Learning-
and-Big-Data-1

Data is the New Oil!
In the last 10 minutes we generated more data than from prehistoric times until 2003!

A data scientist is able to collect, analyze, and interpret data
from a variety of sources (social interaction, business
processes, cyber-physical systems).
Turning data into value!

Four generic data science questions:
1. What happened?
2. Why did it happen?
3. What will happen?
4. What is the best that can happen?

Big data is a broad term for data sets so large or complex that traditional data processing
applications are inadequate.
http://en.wikipedia.org/wiki/Big_data
How Much Data?
1 PB = 1000000000000000B = 1015 bytes = 1000terabytes.
• Google processes 20 PB a day (2008)
• Facebook has 2.5 PB of user data + 15 TB/day (4/2009)
• Each engine of Boeing 747 generates 20 TB of information per hour

Some Big Data Theories and Techniques
Map-Reduce
Pattern Discovery
Social Network Analysis
Process Mining

Social Network
Analysis
(SNA)

Social Network Analysis (SNA)
Every thing is connected
When you sell items …
When you receive customer calls ... When you make a contract …
When you ship orders …

What are Networks?
• Networks are sets of nodes connected by edges
“Network” ≡ “Graph”
node
edge

What is SNA?
SNA (Social Network Analysis) is the mapping and measuring of relationships and
flows between people, groups, organizations, computers, URLs, and other
connected entities
SNA provides both a visual and a
mathematical analysis of human
relationships.

Why do we need Social Network Analysis?
• Are nodes connected through the network?
• How far apart are they?
• Are some nodes more important due to their position in the
network?
• How will be the patterns for information diffusion?
• Is the network composed of communities?

Now,
let’s see some samples of SNA …

Internet
structure of the Internet at the level of autonomous systems. Data source: Mark
Newman http://www-personal.umich.edu/~mejn/netdata/.

Political Blogs
2004 United States Presidential Election Network
Liberals
Conservatives

Facebook Friendship Network

SNA in Organizations (or ONA)

SNA Metrics :
Degree
Betweenness
Closeness

SNA Main Centrality Metrics
Degree
The number of direct connections that a node has
𝑑𝑖 =
𝑗 𝑎𝑖𝑗
(𝑛 − 1)

Betweenness
Betweenness centrality identifies an entity's position within a network in terms of its
ability to make connections to other pairs or groups in a network.
CB (i) = gjk (i)/gjk
j<k
å
CB
'
(i) = CB (i )/[(n -1)(n -2)/2]

Closeness
Closeness centrality measures how quickly an entity can access more entities in a
network.
Cc (i) = d(i, j)
j=1
N
å
é
ë
ê
ê
ù
û
ú
ú
-1
CC
'
(i) = (CC (i))/(N -1)

SNA Tools:
NodeXL

SNA Tools:
Gephi

SNA Tools:
UCINET

Key nodes in Organization
(from ONA view)

Find a node that has high betweenness but
low degree

Find a node that has low betweenness but
high degree
Data Analytics by Vala Ali RohaniData Analytics by Vala Ali Rohani

Process Mining

Process Mining
https://www.coursera.org/course/procmin

Process Mining
Process mining is the missing link between model-based process analysis and
data-oriented analysis techniques.
Process mining seeks the confrontation between event data (i.e., observed behavior)
and process models (hand-made or discovered automatically).
Some example applications include:
• Analyzing treatment processes in hospitals
• Improving customer service processes
• Understanding the browsing behavior of customers using a booking site
• Analyzing failures of a baggage handling system
What is Process Mining?

Process Mining
• What is the process that people really follow?
• Where are the bottlenecks in the studied process?
• Where do people (or machines) deviate from the expected or
idealized process?
• What are the "highways" in my process?
• What factors are influencing a bottleneck?
• Can we predict problems (delay, deviation, risk, etc.) for
running cases?
• Can we recommend some improvements for main process of the
organization?
• How to redesign the process / organization / machine?
Process mining use cases

Process Mining
Some Examples of Real Discovered Processes

Market Basket
Analysis

Introduction
Market Basket Analysis (MBA) is a data mining technique which is widely used in the
consumer package goods (CPG) industry to identify which items are purchased together
and, more importantly, how the purchase of one item affects the likelihood of another
item being purchased.
Bill Qualls, First Analytics, Raleigh, NC, Introduction to Market Basket Analysis

SALES TRANSACTIONS
Our imaginary store sales the following items: bananas, bologna, bread, buns, butter, cereal,
cheese, chips, eggs, hotdogs, mayo, milk, mustard, oranges, pickles, and soda. We have
recorded 20 sales transactions as follows:

MBA Theories:
Support for itemset I = the number of baskets containing all items in I.
Given a support threshold s, sets of items that appear in at least s baskets are called
frequent itemsets.
Association rules are If‐then rules about the contents of baskets.
Confidence of this association rule is the probability of j given i1,…,ik.

MBA Theories:

Market Basket Example

Thank you

Data Analytics

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Data Analytics (20)

Recently uploaded (20)

Data Analytics