Transcript 1/14/16 (Keynote) - Society for Information Management

BUSINESS INTELLIGENCE & ADVANCED ANALYTICS
DISCOVER | PLAN | EXECUTE
JANUARY 14, 2016
Who is...
Data Governance
Master Data
Management
Data Quality
Metadata
Management
Security and
Lifecycle
Management
Data
Integration
Big Data
Hadoop
NoSQL
Spark
Data
Warehouse
Real-Time
Batch
Business
Intelligence
Operational
& Analytical
w w w. i n t r i c i t y. c o m / v i d e o s
Channel
Planning &
Forecasting
Predictive
Analytics
Organizations Diverse Data Needs
CIO Technology Priorities
Business Intelligence Defined
Business Intelligence (BI) provides historical, current
and predictive views of business operations and
supports business decisions ranging from operational to
strategic. Common functions of business intelligence
technologies are reporting, online analytical processing
(OLAP), analytics, data mining, process mining, complex
event processing, business performance management,
benchmarking, text mining, predictive analytics and
prescriptive analytics.
Big Data Supports and Enables BI
Big Data Defined (4 V’s)
Velocity
Volume
Huge data size,
terabytes
- petabytes
Velocity
V
V
V
Variety
Veracity
V
Various data sources (social,
mobile, M2M, structured
and unstructured data)
High speed of data
flow, change
and processing
Veracity
Various levels of data
uncertainty and reliability
Volume
Volume
Velocity
Velocity
Big Data Volume
• increasing data amount
• handling Terabytes – Petabytes
• 1 TB
= 1 000 000 000 000 bytes
= 1012bytes
= 1000 gigabytes
Variety
Veracity
Veracity
• 1 PB = 1 000 000 000 000 000 B
= 1015bytes
= 1000 terabytes
Velocity
Volume
Velocity
Velocity
Velocity of data processing
Need to process data in a fast way,
considering their huge amount (TB, PB)
and variety (distributed, mix of structured
and unstructured data)
E.g.
• Searching Twi er posts
• Fast analysis of video streams (Youtube),
Variety
Veracity
Veracity
• Image recogni on (real- me augmented
reality, online photo faces recogni on),
• Sound analysis (speech recogni on)
• Text analysis (Internet search)
Variety
M2M
Social
media
Mobile
Huge variety of big data
• Data from various data sources
• Social media (Facebook, Twi er)
• Mobile data (loca on, tex ng)
• Machine to Machine systems
(GPS, logs, RFID…)
• Database systems
Velocity
Volume
Velocity
•
Variety
Veracity
Veracity
Handling various data types
• Structured
• records in databases (customers,
billing, stock exchange data…),
Machine logs
• Unstructured
• free text, video, sounds
Veracity
Volume
Variety
Velocity
Velocity
Veracity
Veracity
Veracity of big data
Various data uncertainty and reliability
• Imprecision of data (especially of
unstructured data – e.g. text message can
have double meaning)
• Different level of data quality of structured
data (data values are certain and precise)
and unstructured data (fuzzy interpre ng
of images, free text, speech wording …)
• Technical data quality issues (various
formats, source availability, updates)
Right Information, Right People, Right Time
The right information delivered to the right people at the
right time creates real value by:
•Improve efficiency of operations
•Provide easy access to reports
•Deliver right data at the right time
•Comply with regulatory
•Enable better and faster decisions
 Increase User Adoption &
Productivity
THE
HIGH-PERFORMANCE
 Increase ROI
ENTERPRISE
 Increase confidence in decision
making & compliance
 Improve time to decision making
information repository that
enables the decision makers
in IT & business users and
executives
 Improve human capital management
 Drive accountability
 Accuracy of Information
Big Data circa 1960
Why Big Data?
Governed Information Value Optimization
Why Big Data?
IoT - Fueling the Push to Big Data
Big Data - Lowers Cost
Practical Examples
Fraud
Detection
Compliance
Algorithmic
Trading
Smart
Drugs
Resource
Scheduling
Early
Detection
Churn
Prevention
IoT
Self Driving
Cars
Consumer
Analytics
Demand
Planning
Omni
Channel
One Size Does Not Fit All
Examples
Examples
Small Data
Inventory levels
Characteristics
Characteristics
Hundreds – thousands of
records
Typical
tools
Typical
tools
Analytical
methods
Analytical
methods
Personal computer,
Excel, SPSS, R, other
basic statistics software
Simple statistics
Millions of records,
mostly structured data
Server workstation
computer, Relational
database systems, data
warehouses
Advanced statistics,
business intelligence,
data mining,
Sales records,
Customers database
(small and medium
companies)
Cloud, data centers,
Distributed databases,
NoSQL, Hadoop
MapReduce, Distributed
File Systems, Machine
Learning, Predictive
Analytics
(megabytes)
Customer databases
Large Data
(gigabytes- terabytes)
Big Data (terabytes –
petabytes)
Customer interactions
(social media, mobile),
multimedia (video,
images, free text),
location-based data,
RFIM
Business Intelligence Maturity & Value
Business Intelligence Best Practices
01
Data Governance: Formation, Education & Communication
02
Information Value Optimization
03
Best Practices: Make Good Investments Great (TCO)
04
Organization: Shared Services – Shared Vision (ROI)
05
Support Diverse Business Needs & Systems
Measure Quality to Manage/Enforce Quality – Master Data Management
Move from Hindsight to Insight to Foresight
Use existing technology investment more efficiently – Enterprise Data Warehouse 2.0
Repurpose planned investments to achieve organizational excellence
Aligned People, Process & Technology – Right Skills
www.intricity.com/videos
Arkady Kleyner
Executive VP & Co-Founder
244 Fifth Avenue, Suite 2026
New York, NY 10001
Office: 212-461-1100 x5650
Mobile: 917-434-4783
E-mail: [email protected]