DATA ANALYTICS FOR THE CDO Kirk Borne

Principal Data Scientist and Executive Adviser @KirkDBorne

PRESENTED AT MIT CDOIQ SYMPOSIUM PANEL: “THE FLIP SIDE OF DATA CAPITAL: STRATEGIES FOR ASSESSING DATA LIABILITY AND DATA RISK” Booz Allen Innovation Center, Washington DC

JULY 14, 2017

4 ACTIONS FOR THE CDO IN THEIR FIRST YEAR http://www.informationbuilders.com/about_us/whitepapers/download_form/25791

1) Increase Analytics Availability 2) Transform the Corporate Culture 3) Monetize Your Data 4) Promote Data Governance

… but don’t focus too heavily on data governance, because you may spend your first year doing nothing else. In that case, you won’t have a second year! Booz Allen Hamilton Internal

1

THE 5 LEVELS OF ANALYTICS MATURITY Data Science can be applied at any level, depending on the level of analytics maturity required.

1) Descriptive Analytics

4) Prescriptive Analytics

– Hindsight (What happened?)

– Insight (How can we optimize what happens?) (Follow the dots / connections in the graph!)

2) Diagnostic Analytics – Oversight (real-time / What is happening? Why did it happen?)

3) Predictive Analytics – Foresight (What will happen?)

5) Cognitive Analytics – Right Sight (the 360 view , what is the right question to ask for this set of data in this context = Game of Jeopardy) – Finds the right insight, the right action, the right decision,… right now! = Next Best Action! – Moves beyond simply providing answers, to generating new questions and hypotheses.

http://dilbert.com/strip/2008-05-07 From the Booz Allen “Field Guide to Data Science” Booz Allen Hamilton Internal

2

“ALL THE WORLD IS A GRAPH”

- Shakespeare?

EXAMPLES Connecting the nodes in the graph and labeling the edges (with context and relationship information) can lead to deeper insights than simple transactional databases. • Customer Journey modeling • Data Valuation: links to other data, users, use cases, applications, semantics, reports • Risk Assessment and Data Misuse Mitigation: Accesses vs. Policies - HIPAA, PCI, FERPA, GDPR (EU General Data Protection Regulation), “Right to be forgotten”, …

• Discovery across disconnected document collections, through linked semantic assertions • Causal Factor Analysis: Marketing Attribution, Safety Incident Investigation, …

• Fraud networks, Illegal goods trafficking networks, Money-Laundering networks The connection between black hat entities {1} and {3} never appears explicitly within a transactional database.

{1}

Booz Allen Hamilton Internal

{2}

{3}

3

EVERYTHING IS NOW BEING QUANTIFIED & TRACKED https://mapr.com/blog/big-data-everything-quantified-and-tracked-what-means-you/

• In the Big Data era, everything is quantified and tracked! • Examples: • • • • • • •

Social Networks Population & Personal Health Smart Cities & Smart Highways Retail & Marketing Analytics Cybersecurity Sustainability Development Goals … The Internet of Things

• The Internet used to be a thing, now things are the Internet. • Knowing the knowable via deep, wide, fast data from ubiquitous sensors! • The Internet of Everything!

• Put those data to use! … as evidence for your valuation and risk analyses (assessment and mitigation) Booz Allen Hamilton Internal

4

Use Data Analytics to Monitor Compliance with GAPP (Generally Accepted Privacy Principles) http://www.lexisnexis.com/privacy/data-privacy-principles.aspx

• • • • • • • • •

Security (information will be secured by firewalls, passwords, …) Protection of PII (extra conditions are placed on who can see it) Accuracy (keepers must strive to keep it accurate and up-to-date) Protection of key ID numbers (Driver’s License, Social Security) Reputable sources (legitimate inquiries to legitimate sources) Choice & Consent (to opt in or opt out) Access & Correction (you have the right to check & fix errors) Accountability (keepers of the data must take responsibility for it) Online privacy (special precautions are needed – e.g., security questions, encryption, multi-factor authentication) • Disclosure (only under precisely specified conditions or agreements) • Compliance (with applicable laws and regulations) • Monitoring (both the security of the information & compliance) Booz Allen Hamilton Internal

5

THANK YOU!

Booz | Allen | Hamilton

Learn how AI and Machine Intelligence empower The Mathematical Corporation …

LISTEN @KirkDBorne @BoozDataScience

READ, BUILD, and EXPLORE www.boozallen.com/datascience      

Tips for Building a Data Science Capability The Mathematical Corporation 10 Signs of Data Science Maturity The Field Guide to Data Science The Data and Analytics Catalyst Explore: sailfish.boozallen.com

PARTICIPATE datasciencebowl.com

Booz Allen Hamilton Internal

6