Big data-analytics-changing-way-organizations-conducting-business

1
© 2015 MetricStream, Inc. All Rights Reserved.
Big Data and Analytics: Changing the Way Organizations
are Conducting Business
Dr. Kirk D. Borne
Data Scientist & Advisor
Big Data Consultant
Vibhav Agarwal
Sr. Manager of Product Marketing
MetricStream

2
Today’s Agenda
 Data science for driving business innovation
 Knowledge discovery and data mining systems for better governance
 Analytics automation for just-in-time insights for mitigating risks
 Decision science-as-a-service for marketing, retail, financial, security, and other sectors
 Question & Answer

Big Data and the fundamental
business conflict:
RISK versus REWARD
http://www.telegraph.co.uk/news/worldnews/europe/russia/10061780/Russian-convicts-beat-Americans-in-cyber-chess-battle.html

Challenges in Digital Business
Demands on Data Analysts
Multiple data sources, stakeholders, and
constituencies require business analysts to extract
insights across a variety of digital user communities
and portals (internal employee self-service, internal
IT and cybersecurity systems, your business
customer interaction channels, B2B “customers”).
5
Lack of True Automation
Lack of automated processes prevents analysts from achieving targeted
end-user digital content delivery and performing in-depth analytics on
massive digital data streams.
© SYNTASA 2014

Big Data: What is it good for?
The 3 D2D’s
 Knowledge Discovery
– Data-to-Discovery (knowledge insights)
 Data-driven Decision Support
– Data-to-Decisions (decisioning insights)
 Big ROI (Return On Innovation) !!!
– Data-to-Dividends (innovation insights)
– Data-to-Dollars (business ROI)
6

1) Correlation Discovery
 Finding patterns, trends, and dependencies, which
might reveal new principles of behavior
2) Novelty Discovery
 Finding new, rare, one-in-a-[million / billion / trillion]
objects and events
3) Class Discovery
 Finding new classes of objects, events, and behaviors
 Learning the rules that constrain class boundaries
4) Association Discovery
 Finding unusual (improbable) co-occurring associations
Data Science in 4 easy steps
(achieving the 3 D2D’s from your Big Data)

4 Business Examples:
Association Discovery
(for recommender engines)

 Classic Textbook Example of Data Mining (Legend?):
Data mining of grocery store logs indicated that men who
buy diapers also tend to buy beer at the same time.
Business Example #1

 Amazon.com mines its customers’ purchase logs to
recommend books to you: “People who bought this book
also bought this other one.”
Business Example #2

 Netflix mines its video rental history database to
recommend rentals to you based upon other customers
who rented similar movies as you.
Business Example #3

 Wal-Mart studied product sales in their Florida stores in
2004 when several hurricanes passed through Florida.
 Wal-Mart found that, before the hurricanes arrived, people
purchased 7 times as many of {one particular product}
compared to everything else.
Business Example #4

 Wal-Mart studied product sales in their Florida stores in
2004 when several hurricanes passed through Florida.
 Wal-Mart found that, before the hurricanes arrived, people
purchased 7 times as many strawberry pop tarts
compared to everything else.
Business Example #4

Strawberry pop tarts???
http://www.nytimes.com/2004/11/14/business/yourmoney/14wal.html
http://www.hurricaneville.com/pop_tarts.html
http://bit.ly/1gHZddA

Knowledge Discovery for multi-source Data:
Heterogeneous data collections are the new normal
New Knowledge on
correlations, causal
connections, and
interdependencies
between events,
objects, processes
within any
application domain
Data to Information to Knowledge

New Knowledge on
connections, and
interdependencies
between events,
objects, processes
within any
application domain
The “first mile” challenge:
integrating multi-source data
The “first mile” challenge:
integrating multi-source data

New Knowledge on
connections, and
interdependencies
between events,
objects, processes
within any
application domain
The “last mile” challenge:
deriving Actionable Intelligence
from all of your data sources.

The MIPS model
for Dynamic Data-Driven Application Systems (DDDAS)
• MIPS =
– Measurement – Inference – Prediction – Steering
• This applies to any Network of Sensors:
– Web user interactions & actions (web analytics data), Cyber network
usage logs, Social network sentiment, Machine logs (of any kind),
Manufacturing sensors, Health & Epidemic monitoring systems, Financial
transactions, National Security, Utilities and Energy, Remote Sensing,
Tsunami warnings, Weather/Climate events, Astronomical sky events, …
• Machine Learning enables the “IP” part of MIPS:
– Autonomous (or semi-autonomous) Classification
– Intelligent Data Understanding
– Rule-based
– Model-based
– Neural Networks
– Markov Models
– Bayes Inference Engines
Alert & Response systems:
• Actionable insights from
streaming business data
• Automation of any data-
driven operational system
http://dddas.org

The MIPS model
for Dynamic Data-Driven Application Systems (DDDAS)
• MIPS =
– Measurement – Inference – Prediction – Steering
• This applies to any Network of Sensors:
– Web user interactions & actions (web analytics data), Cyber network
usage logs, Social network sentiment, Machine logs (of any kind),
Manufacturing sensors, Health & Epidemic monitoring systems, Financial
transactions, National Security, Utilities and Energy, Remote Sensing,
Tsunami warnings, Weather/Climate events, Astronomical sky events, …
• Machine Learning enables the “IP” part of MIPS:
– Autonomous (or semi-autonomous) Classification
– Intelligent Data Understanding
– Rule-based
– Model-based
– Neural Networks
– Markov Models
– Bayes Inference Engines
http://dddas.org
Alert & Response systems:
• Actionable insights from
streaming business data
• Automation of any data-
driven operational system

From Sensors to Sentinels to Sense:
Take Data to Information to Knowledge
to Insights (and Action!)
 From Sensors (Measurement & Data Collection)…
 … to Sentinels (Monitoring & Alerts) …
 … to Sense-making (Data Science) …
 … to Cents-making (Business ROI)
… Actionizing and Productizing Big Data
21

Smart Engines for Data-Driven
Discovery and Decision Support
• New knowledge and insights are acquired by mining
actionable data from all digital inputs (Sensors!)
• Decisions are based on the new knowledge mined,
prior experience, and your “business” decisioning rules
embedded within the pipeline (Sentinels!)
• “Smart Sensors” act autonomously in real-time, without
human intervention = actionable intelligence (Sense!)
http://legacy.samsi.info/200506/astro/presentations/tut1loredo-7.pdf 22

Decision Analytics – based on massive amounts of information
(Big Data – What is it good for? …Decision Support and Innovation!)
From Devices……
… Intentions…
… Location, weather, and
other geographic attributes…
… Demographics…
23

Automating
Analytics
as-as-Service
(AaaS)
• Based on SYNTASA’s Marketing Analytics-as-a-ServiceTM
(MAaaS)
• “Smart Sentinel in a box”
– Your business rules determine the goals, decision points, alerts, and responses.
– Moving beyond historical hindsight and oversight (Descriptive & Diagnostic
Analytics) to new world of insight and foresight (Predictive & Prescriptive AaaS),
eventually achieving right sight (Cognitive Analytics = the 360 view, enabling the
right action, for the right web user, at the right place, at the right time).
• Mining multi-portal big data streams (across the organization’s departments)
• Personalization and Customization (“segment of one”)
• Decision Automation in a rich content (Big Data) environment
24Based on Marketing Analytics-as-a-ServiceTM
(MAaaS) from http://www.syntasa.com/
Digital user
Behavior
Modeling

The New Digital Business:
Big Data Analytics Challenge = Risk Mitigation
• General example of streaming data analytics:
 Real-Time Event Mining for Actionable Intelligence:
 Identifying, characterizing, & responding to millions of events in real-time streaming data
 Deciding which events (out of millions) need investigation and/or response
• Web Analytics example:
 Web Behavior Modeling and Automated System Response (from
online interactions & web browse patterns, personalization, user
segmentation, 1-to-1 marketing, advanced analytics discovery,…)
• Many other examples:
 Health alerts (from EHRs and national health systems)
 Tsunami alerts (from geo sensors everywhere)
 Cybersecurity alerts (from network logs)
 Social event alerts or early warnings (from social media)
 Preventive Fraud alerts (from financial applications)
 Predictive Maintenance alerts (from machine / engine sensors)
RiskMitigation

The New Digital Business:
Big Data Analytics Rewards = Innovation & Value
• Learning from Data (Data Science)…
– Clustering (= New Class discovery, Segmentation)
– Correlation & Association discovery
– Classification, Diagnosis, Prediction
– Outlier / Anomaly / Novelty / Surprise detection
• … to conquer the 3 D2D challenges:
– Data-to-Discoveries
– Data-to-Decisions
– Data-to-Dividends
(big ROI = Return on Innovation)
26
Rewards!

27
Leveraging Big Data Analytics in GRC
Vibhav Agarwal
MetricStream

28
From Integrated to Pervasive GRC
Widespread and rapid
adoption of new
technologies (e.g.,
mobile, social)
Increasing regulatory
pressures and Board /
Management
accountability
Represents internally developed solutions Represents vendor solutions
First Generation Second Generation Third Generation Fourth Generation
ExpandingGRCApplications
2003 2013 ?
Sarbanes-Oxley (SOX) enacted following series of
accounting scandals (Enron, Tyco, WorldCom) Global financial crisis
Siloes lead to greater
risk and inefficient use
of resources
Audit / Finance
(Sarbanes-Oxley)
Audit / Finance
IT GRC
Audit / Finance
IT GRC
Legal
Quality
Management
Compliance
Management
IT GRC
Legal
Quality
Management
Notable disasters include Deepwater Horizon and Fukushima
Risk
Management
Standalone, largely
ad hoc, internally
developed solutions
Siloed vendor and
internally developed
point solutions
Integrated GRC
platform solutions
Pervasive GRC
Audit / Finance
IT GRC
Legal
Quality
Management
Compliance
Management
Risk
Management
Vendor Risk
Management
Social GRC
Long-Tail Apps
Comprehensive&UnifiedAnalytics
Cloud GRC
CommonDataModel;CustomizablePlatform

29
Big Data: Imperative for Pervasive GRC

30
5 Mega Trends Driving this Big Data requirements
Globalization – Explosion of rules, policies, data, and
regulations as organizations extend across countries
Virtualization – Transfer of critical data on cloud for
scalability and efficiency to drive the TCO of IT systems
lower
Mobility – Ubiquitous Access to data across devices for
employees, customers and partners
Social Media – New set of imperfect data for Real time
approximate Risk intelligence. Extensive sharing of internal
data. Blurring of traditional organization boundaries
Hyper-Connectivity – Expansion of employee, vendor
and supply chain ecosystem into a real-time collaborative
network

31
GRC: A Big Data Problem
Multiple GRC Data Sources, Event Co-relations
Content and Standards Library ERP, SCM, Content Management applications
Network Frontiers/UCF, NIST NVD, Cloud Security
Alliance, SharedAssessments.org
SAP, Oracle, i2, Ariba, JD Edwards, EMC,
Documentum, OpenText, Sharepoint
Threat , Vulnerability, Logs, SIEMS, Operations and Asset Management
nCircle, Nessus , Qualys, Symantec, McAfee,
Arcsight, Splunk, BigFix, eEye
HP Asset Manager, BMC Remedy
SIEM, Log Management, Application Intelligence Risk Models
LogLogic, ArcSight, Splunk Market and Credit Risk Models, RiskMetrics, RMA
Segregation of Duties, CCM, Transaction
Monitoring
Risk and Framework Content
CrossIdeas , Engiweb Security, Greenlight, Mantaz,
Actimize, MES systems
ORX, Gold; American Banking, OCEG, IIA, ISO, D&B,
Configuration Management Regulatory Content sources
Qualys, nCircle Configuration Compliance Manager
(CCM), eEye Retina CS
Lexis, Factiva, Complinet, Reuters, FDA, State Regs
ComplianceOnline - > 1000 sources
Data Loss, EndPoint, Mobile, Application Security Smart Grid and Green Data centers
Verdasys, Sophos, Veracode, Lookout, Symantec Cisco, SilverSpring
Social Media Sources News Feeds

32
Big Data: Solving the Key Challenge
How to channelize the data
to right stakeholder?
How can the situation be
mitigated in real-time?
How to filter Voice from
Noise in the Social Media?
•Hadoop DFS based framework to allow aggregation of content
across data sources
•Ability to handle both structured and unstructured content
Aggregate data across Social
Media sources
•Advanced text analytics based on custom rules to identify text
patterns and indicators of risk.
•Sentiment analysis and scoring mechanism to prioritize the
identified data.
Advanced Text analytics for
Sentiment Identification
•Create custom dashboards and workflows to channelize the
information to right stakeholder.
•Identify any risk or gap in the content and channelize through
custom workflow.
Configuration of custom
workflow and dashboards

33
Big Data: A Effective Risk Management Tool
Trends predict
Super Cyclone in
India
90% of
Manufacturing
Plants impacted
No supply till plants
restored
Anticipate, Counter
supply disruption
with remedial plan
and publish it
Stock stable
Super Cyclone in
India
90% of
Manufacturing
Plants impacted
No supply till plants
restored
News of disruption
in supply
Stock volatile
10.13 10.30
10.35
14.35
10.10 10.30
10.35Next
day

34
Situational Awareness for BCP
• Track Social Media platforms like:
─ Twitter
─ Facebook
─ Pinterest
─ Google (Google +, Youtube, Crisis Map etc.)
• Correlate Information with Organizational Assets /
Facilities / Risks
• Trigger / Update Incident Management Workflows &
Notifications
• Real-Time Reports &
Dashboards
• Leverage Social Media for
Communications During
Emergencies

35
Big Data Risk Analysis – A Product Reputation Use Case
Social Media site
Postings
Call center transcripts Customer Support
Emails Internal data & reports
Identify the key data
sources for gathering the
Product reputation and
quality feedback
Aggregate & Process the
data using Hadoop DFS
and MapReduce framework
Detect the risks using
natural language processing
based rules, keywords and
author profiles and influence
Inform the relevant
stakeholders through trend
analysis reports and
dashboards
Hadoop DFS
Store the complete data in a Distributed File system
Create risk detection
rules based on key
words, repetition
frequency & Author
influence
Analyze the product
feedback data based on
the rules on a real-time
basis
Reduce the data to highlight
the key product & brand
reputation risks and their
causes
Create trend analysis dashboards to highlight key product feedback categories and risks and
causes highlighted based on the analysis

36
Big Data Risk Analysis– Vendor Due Diligence Use Case
Big Data
Analytics
Unstructured data sets :
News feeds, Social
Media comments
External databases:
Exports registry, PEP
Database ,
Rating Agency Databases
Internal databases:
Vendor information,
Credit and Payment
information
 Aggregate Real time and Up-to-date Vendor Due diligence and Assessment
information
 Correlate the vendor data against key identified risks for accurate risk scoring
and assessment
 Manage compliance to FCPA, UK Bribery Act & OECD Convention etc.

37
Big Data Risk Analysis– IT-GRC Use Case
Aggregate the vulnerability
bulletins across websites e.g.
www.xssed.com, www.iss.net etc…
Analyze the feeds based on the
text analytics based rules and IT
Asset library
Highlight the risks & vulnerabilities
based on the asset library as well
as the rules engine Correlate the Product and CVE details with
the internal IT asset libraries and highlight
potential risks and vulnerabilities

38
Correlate & Improve Product Information
Aggregate the product
information across websites
Analyze the feeds using text
analytics to look to Text Patterns
Highlight any risks & issues based
on the patterns and correlation
with internal databases
Aggregate
Analyze
Correlate

39
About MetricStream
Vision
Integrated Governance, Risk and Compliance for Better Business
Performance
Solutions
• Policy & Compliance Management
• Risk Management
• Business Continuity Management
• IT GRC
• Audit Management
• Supplier Governance
• Quality Management
• EHS & Sustainability
• Governance & Ethics
• Content and Training
• Over 1,400+ employees
• Headquarters in Palo Alto, California with offices worldwide
• Over 350 enterprise customers
• Privately held – Goldman Sachs minority owner
Differentiators
• Technology - GRC Platform – 9 Patents
• Breadth of Solutions – Single Vendor for all GRC needs
• Cross-industry Best Practices and Domain Knowledge
• ComplianceOnline.com - Largest Compliance Portal on the Web
Organization

40
Q&A
Please submit your questions to the host by typing into the chat box on
the lower right-hand portion of your screen.
Thank you for participating!
A copy of this presentation will be made available to all participants in next 48 working hours.
For more details on upcoming MetricStream webinars: http://www.metricstream.com/webinars/index.htm
Dr. Kirk D. Borne
Data Scientist & Advisor
Big Data Consultant
Email: kborne@gmu.edu
Vibhav Agarwal
MetricStream
Email: vibhav.agarwal@metricstream.com

41
Thank You
Contact Us:
Website: www.metricstream.com | Email: webinar@metricstream.com
Phone: USA +1-650-620-2955 | UAE +971-5072-17139 | UK +44-203-318-8554

Big data-analytics-changing-way-organizations-conducting-business

Empfohlen

Empfohlen

Weitere ähnliche Inhalte

Was ist angesagt?

Was ist angesagt? (20)

Andere mochten auch

Andere mochten auch (10)

Ähnlich wie Big data-analytics-changing-way-organizations-conducting-business

Ähnlich wie Big data-analytics-changing-way-organizations-conducting-business (20)

Kürzlich hochgeladen

Kürzlich hochgeladen (20)

Big data-analytics-changing-way-organizations-conducting-business