Documente Academic
Documente Profesional
Documente Cultură
1
Leading the way in Business Solutions
Revenue
Enhancement
Insights from the machine data
Leveraging Flutura IP to build a
M2M Analytics platform
Monetize the M2M data
Leading Healthcare Trust in
UK
Analytics Platform for a
leading NHS Trust
Focused on tracking patient
journey, effectiveness of
treatment, acting on
exceptions and compliance
Healthcare
Analytics
Voice of Customer
Leading Online
Travel Agency
1 TB per month
Search to Booking Ratio
improvement
Personalized content
recommendation
Leading Automotive
Manufacturer
Customer contact center
transcripts + structured data
Sentiment analysis of call
center interactions
Feedback to allied
departments like R&D,
Quality and Process
Engineering
Leading Telecom in Middle
East
300 TB of data from
plethora of devices
Multiple variety of switch,
router, failure events on
network
Detect and mitigate risk and
threat events
Monitor and optimize
capacity utilization
Regulatory compliance
Compliance
Want to begin the journey to become a
Data Scientist ?
3
BIG DATA !!!!!
Operational Analytics will Lead - The Big Data Promise!!!
RFID
Telecom Switches
GP/Telematics
Medical sensors
Tower Data
Auto Sensors
Search Logs
Shopping Basket
1. Online Consumer Activity
2. Device / Sensor Proliferation
3. Data Variety & Velocity
6. Cost of Storage
4. Democratization of tools
5. Machine learning Capabilities
Forces at play
Assessments of Big Data Market
( Source = Mc Kinsey + Gartner )
still businesses are grappling to determine
dollar denting use cases
Large amount of data is being generated
managing data is becoming possible
through new tools
Data will grow 80%
over the next 5 years
and 80% will be
unstructured
- Gartner
Volume + Velocity + Variety = New Possibilities
Data Scientist Sexiest Job of the 21
st
Century
Data
Scientist
What is Social Network Analysis
SNA is the methodical analysis of social network
SNA views social relationships in term of network theory
consisting of nodes and edges
Concepts
Network
Networks are all around us. Like
Public transportation, connecting different place
Power grids, that supplies energy
Facebook Network, how people are connected
Mathematicians describe a network as a collection of
elements and the connections between them.
Nodes and Edges
Nodes: Representing individual actors within network
Node Attribute:
From immediate network: Degree
From entire graph: Closeness, betweeness, centrality, etc.
Edges: Relationship between the individuals
Directed and Undirected:
Directed: A likes B, A to B, etc
Undirected: A and B are friends, etc
Edge Attributes:
Weight (frequency of communication)
Ranking (Best friend and second best friend)
Type (friend, relative and co-worker)
B
C
D
A
Degree
In-Degree: Directed edges that incident on an node
Out-Degree: Directed edges that originate from a node
Degree: In case of undirected network it is just degree. Its the
number of edges that incident on an node
B
C
D
A
Centrality
Based on Betweeness
How many pairs of individuals(other nodes in the network) would have
to go through you(node for which it is calculated) in order to reach one
another in the minimum number of hops?
Example:
C
B
(i) g
jk
(i) / g
jk
j<k
Where g
jk
= the number of shortest paths connecting jk
g
jk
(i) = the number that actor i is on.
A B C E D
Centrality
Based on Closeness
How central you are depends on the length of the average shortest path
between a node and all other nodes in the network
C
c
(i) d(i, j)
j 1
N
1
A B C E D
C
c
'
(A)
d(A, j)
j1
N
N 1
1+ 2+3+ 4
4
10
4
1
0.4
Centrality
Based on Eigen Vector
How central you are depends on how central your neighbors are.
Clustering co-efficient
The clustering coefficient, is a measure of extent to
which nodes in a graph tend to cluster together
The global clustering coefficient is based on triplets of
nodes
A triplet consists of three nodes that are connected by either two (open
triplet) or three (closed triplet) undirected ties. A triangle consists of
three closed triplets, one centred on each of the nodes. The global
clustering coefficient is the number of closed triplets (or 3 x triangles)
over the total number of triplets (both open and closed)
Use-Cases
Identify barrier for internal communication.
Team Building
Identify users with the most influence, and
hence spotting them as potential targets for
marketing communication
Marketing
Identify potential candidates best suited for
a particular position
Human
Resources
Concept
Discovery of insights from a large amount of different
unstructured textual resources.
Deriving high-quality information from text.
Devising of patterns and trends through means such as
statistical pattern learning.
Data
Mining
Text
Mining
Data
Retrieval
Information
Retrieval
Search
(goal-oriented)
Discover
(opportunistic)
Structured
Data
Unstructured
Data (Text)
Use-cases
Major issues faced by the customers
Is the any seasonality in the issues faced
Is there any issues highly present in a particular
location
Customer Service
Industry
Based on the customers email, identify the
customers issue and forward the email to the
most relevant person.
Based on historic data identify what can be
potential issue that could be faced by the
customer and hence improve the services
Advanced Contextual
Text Mining
Affinity & Sequence Analysis
Concept
Affinity analysis technique that discovers co-occurrence relationships
among activities performed by (or recorded about) specific individuals or
groups
Sequence mining is a topic of data mining concerned with finding
statistically relevant patterns between data examples where the values are
delivered in a sequence.
Use-Cases
Purchase behavior of a customer
Products purchased together
Market Basket
Analysis
Order in which the consumer buys
Buying Pattern in
the Consumer
Normal flow in a website
Most frequented webpage
Path used to reach this page
Browsing Pattern
in a Website
Recommendation Engine
Working
Use-Cases
Recommendation for the user on
products they might like, e.g Amazon,
ecommerce
Suggesting the precaution one has to
take
Health-care
The machines that might fail because
of particular reason
Manufacturing
Nothing delights us more than winning your
Trust!!!
32
Team
Alen Sebastian-alen.sebastian@flutura.com LinkedIn Profile
Jobil Louis-jobillouis.joseph@flutura.com LinkedIn Profile
Samir Madhavan-samir.madhavan@flutura.com LinkedIn Profile
Sharan Kumar R-sharankumar.r@flutura.com LinkedIn Profile
Thank You