A workshop on 'AI transforming Business’

© DataMites™ | Feb 2019 | CII – Confederation of Indian Industry | AI Transforming Business
Artificial Intelligence (AI) Transforming
Business
Two Day Workshop for business leaders, 20-21st Feb 2019, Chennai
Hosted by Confederation of Indian Industry (CII)
Facilitated by DataMites™
Hosted by: Sponsored by:

• 17+ years in leading innovation projects and bringing business
Ideas to working real world solutions. Currently involved in
applied AI solutions to Business as CEO of Rubixe™
• Coaching and mentoring 2000+ AI professionals and leading
corporates in India
• 2007-2014 : As management consultant based in Netherlands,
worked with senior management of top companies in Europe
in shaping Tech-based Innovation propositions
• Academia
• Engineering in Electronics from NITW, India
• MBA from University of Amsterdam – Netherlands &
IIMA, India.
• Ph.D Scholar – Researching on AI in consumer behavior
Ashok Kumar Adinarayanan
AI Evangelist, Entrepreneur, Ph.D Scholar, MBA
Workshop Speaker : AI Transforming Business

Evolution of Artificial Intelligence
(AI)

Life dates back to billions of years..
Humans got smarter in just 10,000 years.
TheNextStep - Data Science Workshop 4

THE DAWN OF
ARTIFICIAL INTELLIGENCE

Artificial Intelligence
AI is typically defined as the ability
of a machine to perform cognitive
functions we associate with human
minds, such as perceiving,
reasoning, learning, interacting with
the environment, problem solving,
and even exercising creativity.

Al-Jazari’s programmable automata
(1206 CE)
Source: www.historyofinformation.com

A Brief History of Artificial Intelligence
• 1950 - Alan Turing released a paper titled “Computing Machinery and
Intelligence.” It became famous and now known as “Turing Test”
• 1956 - cognitive scientist John McCarthy introduced the term “artificial
intelligence” at a summer workshop known as the Dartmouth Conference. The
extended brainstorming session lasted roughly eight weeks
• 1970s and ’80s - AI Winter, Minsky and AI theorist Roger Schank warned
business community that AI would eventually lead to disappointment
• 1986 - Even as investors fled, keen academics and researchers kept forging
ahead. Geoffrey Hinton, a professor at Carnegie Mellon University, described a
new learning procedure—the back-propagation algorithm
• The 1990s And Beyond: AI And Machine Learning’s Eternal Spring. Andrew Ng

AI Today

Why Artificial Intelligence now?
• That brings us to today. In the title of his talk was fitting: “Artificial
Intelligence Is The New Electricity.” Andrew Ng cites three reasons
why AI now:
1. The availability of big data
2. Supercomputing Power
3. Advancements in Modern algorithms

John McCarthy. , Coined Ai in 1955
Known as Father of Artificial Intelligence
Worked mostly for Stanford University.
Yann Lecun
Chief Artificial Intelligence
Scientist at Facebook AI
ResearchGeoffrey Hinton
Known as God Father of Deep
Learning. Chief Data Science @
Google.
16
Andrew Ng
Ng co-founded and led Google
Brain and was a former VP &
Chief Scientist at Baidu

Areas of Artificial Intelligence
17
Computer Vision
Natural Language Processing
Machine Learning & Deep Learning Decision Making
Robotics

Investment Made AI
Categories
Machine Learning
received most investment
$ 5 to 7 billion

Machine Learning

What is Machine Learning?
Machine learning gives
computers the ability to
learn without being explicitly
programmed

Amazon Returns Policy Case Study
Amazon Return Policy Exploitation
A Machine Learning Case Study - Demo

Types of Machine Learning
• Supervised Machine Learning
• Unsupervised Machine Learning
• Reinforcement Learning
• Transfer Learning

Supervised Machine Learning
What is it?
An algorithm uses training data and feedback
from humans to learn the relationship of given
inputs to a given output (eg, how the inputs
“area” and “location” predict housing prices)
When to use it?
You know how to classify the input data and
the type of behaviour you want to predict, but
you need the algorithm to calculate it for you
on new data
How it Works?
A human labels the input data (eg, in the case of
predicting housing prices, labels the input data as
“area,” “location” etc) and defines the output variable
(eg, housing prices). The algorithm is trained on the data
to find the connection between the input variables and
the output. Once training is complete–typically when
the algorithm is sufficiently accurate–the algorithm is
applied to new data

Supervised Machine Learning
Regression Models Classification Models
• Linear Regression
• Lasso & Ridge Regression
• Polynomial Regression
• ElasticNet Regression
• Support Vector Classifier
• Naïve Bayes Classifer
• Random Forest Classifier
• XGBoost Classifier

Unsupervised Machine Learning
What is it?
An algorithm explores input data without
being given an explicit output variable (eg,
explores customer demographic data to
identify patterns)
When to use it?
You do not know how to classify the data, and
you want the algorithm to find patterns and cl
assify the data for you
How it Works?
The algorithm receives unlabeled data (eg, a set of data
describing customer journeys on a website)
It infers a structure from the data
The algorithm identifies groups of data that exhibit simil
ar behaviour (eg., forms clusters of customers that exhib
it similar buying behaviours)

Reinforcement Learning
• Reinforcement learning is an important
type of Machine Learning where an
agent learn how to behave in a
environment by performing actions and
seeing the results.
• Simply learning from mistakes and aim
to maximize rewards

Transfer Learning
• Transfer learning is a research
problem in machine learning that
focuses on storing knowledge
gained while solving one problem
and applying it to a different but
related problem
• This is natural to humans. eg., if you
learn to drive bicycle, you can learn
motorbike relative easy.

Deep Learning Introduction

Deep learning is a type of machine learning that can process a wider range of data resources, requires less data pre-
processing by humans, and can often produce more accurate results than traditional machine-learning approaches
(although it requires a larger amount of data to do so). In deep learning, interconnected layers of software-based
calculators known as “neurons” form a neural network. The network can ingest vast amounts of input data and
process them through multiple layers that learn increasingly complex features of the data at each layer. The network
can then make a determination about the data, learn if its determination is correct, and use what it has learned to
make determinations about new data. For example, once it learns what an object looks like, it can recognize the
object in a new image.

CNN
• What?
• A multi-layered neural network with a special architecture designed to extract increasingly
complex features of the data at each layer to determine the output
• When?
• When you have an unstructured data set (eg, images) and you need to infer information from it

Deep Learning CNN Use-Cases

RNN
• What?
• A multi-layered neural network that can store information in context nodes,
allowing it to learn data sequences and output a number or another
sequence
• When?
• When you are working with time-series data or sequences (eg, audio
recordings or text)

Convolution Neural Network – CNN Demo
Classification of images – Cats & Dogs

AI and ML Applications

https://hbr.org/2017/05/how-machine-learning-is-
helping-us-predict-heart-disease-and-diabetes
https://futurism.com/confirmed-ai-can-predict-heart-
attacks-and-strokes-more-accurately-than-doctors/
AI Predict Heart Attack
Better Than Doctors
University of Nottingham researchers
created an AI system that scanned
routine medical data to predict which
patients would have strokes or heart
attacks within 10 years.
Health Care

http://policestate.news/2017-06-08-uk-police-will-start-
using-ai-to-assign-threat-levels-to-criminals-ultimately-
deciding-how-long-suspects-should-be-kept-in-custody.html
Police officials in Durham, U.K. are slated to roll
out an artificial intelligence system designed to
help the authorities determine whether or not a
suspect should be kept in police custody.The
system, known as the Harm Assessment Risk Tool
(HART) will classify suspects as low, medium, or
high risk offenders.
Finding Crinimals
Law and Order

https://www.forbes.com/sites/bobevans1/2017/06/20/how-
google-and-amazon-are-torpedoing-the-retail-industry-with-
data-ai-and-advertising/#766d80cb5c66
http://www.mckinsey.com/business-functions/mckinsey-
analytics/our-insights/the-age-of-analytics-competing-in-a-
data-driven-world
what to stock, how much to buy, what products to
suggest to repeat customers. But doing more with
that data using machine learning is just what
retailers need to really succeed in the current
market.
Machine Learning in Retail
Retail

AI in Banking
• Fraud Transactions
Detection
• Predicting risk of asset
class based on future
context models.
• More personalized and
faster customer
experiences

AI will fight challenging diseases
We'll Soon Trust AI
More Than Doctors
From Diagnosis to Treatment

Skin Cancer Deduction
Facial Recognition
HR Analytics
Recommendations
more..

A Good Artificial Intelligence
Use case

Good AI/ML Use Case should have consider
below points.
• Is the data needed for the use case accessible and of sufficient quality
and time horizon?
• Is it clear, what specific process steps would need to change for a
particular use case?
• Is the team involved in that process have to change?
• What could be changed with minimal disruption, and what would
require parallel processes until the new analytics approach was
proven?

AI Use Cases in Practice

Financial Services - Experian
• With approximately 3.6 petabytes of data (and
growing) about individuals around the world,
credit reference agency Experian gets its
extraordinary amount of data from marketing
databases, transactional records and public
information records.
• They are actively embedding machine learning
into their products to allow for quicker and more
effective decision-making. Over time, the
machines can learn to distinguish what data
points are important from those that aren’t.
Insight extracted from the machines will allow
Experian to optimize its processes.

Financial Services - American Express
• American Express processes $1 trillion in
transaction and has 110 million AmEx cards in
operation.
• They rely heavily on data analytics and machine
learning algorithms to help detect fraud in near
real time, therefore saving millions in losses.
• Additionally, AmEx is leveraging its data flows to
develop apps that can connect a cardholder with
products or services and special offers.
• They are also giving merchants online business
trend analysis and industry peer benchmarking.

Healthcare - Infervision
• AI and deep learning is being put to use to save
lives by Infervision.
• In China, where there aren’t enough
radiologists to keep up with the demand of
reviewing 1.4 billion CT scans each year to look
for early signs of lung cancer.
• Radiologists need to review hundreds of scans
each day which is not only tedious, but human
fatigue can lead to errors.
• Infervision trained and taught algorithms to
augment the work of radiologists to allow them
to diagnose cancer more accurately and
efficiently.

Manufacturing - Volvo
• Cars are increasingly connected and
generate data that can be used in a number
of ways.
• Volvo uses data to help predict when parts
would fail or when vehicles need servicing,
uphold its impressive safety record by
monitoring vehicle performance during
hazardous situations and to improve driver
and passenger convenience.
• Volvo is also conducting its own research
and development on autonomous vehicles.

Manufacturing - BMW
• BMW has big data-related
technology at the heart of its
business model and data guides
decisions throughout the business
from design and engineering to
sales and aftercare.
• The company is also a leader in
driverless technology and plans for
its cars to deliver Level 5
autonomy—the vehicle can drive
itself without any human
intervention—by 2021

Media - BBC
• The BBC project, Talking with Machines is
an audio drama that allows listeners to
join in and have a two-way conversation
via their smart speaker.
• Listeners get to be a part of the story as it
prompts them to answer questions and
insert their own lines into the story.
• Created specifically for smart speakers
Amazon Echo and Google Home, the BBC
expects to expand to other voice-
activated devices in the future.

Media - Netflix
• Big data analytics is helping Netflix
predict what its customers will enjoy
watching.
• They are also increasingly a content
creator, not just a distributor, and use
data to drive what content it will invest in
creating.
• Due to the confidence they have in the
data findings, they are willing to buck
convention and commission multiple
seasons of a new show rather than just a
pilot episode.

Retail - Burberry
• When you first think of Burberry, you likely
consider its luxury fashion and not first
consider them a digital business.
• However, they have been busy reinventing
themselves and use big data and AI to
combat counterfeit products and improve
sales and customer relationships.
• The company’s strategy for increasing sales is
to nurture deep, personal connections with
its customers.
• As part of that, they have reward and loyalty
programs that create data to help them
personalize the shopping experience for each
customer

Retail - Walmart
• As the world’s second-largest retailer, Walmart is
on the cutting edge of finding ways to transform
retail and provide better service to its customers.
• They use big data, machine learning, AI and the
IoT to ensure a seamless experience between the
online customer experience and the in-store
experience (with 11,000 brick-and-mortar stores)
something rival Amazon isn’t able to do.
• Enhancements include using the Scan and Go
feature on the app, Pick-up Towers and they are
experimenting with facial recognition technology
to determine if customers are happy or sad.

Service - Microsoft
• Central to everything Microsoft does is leveraging
smart machines.
• Microsoft has Cortana, a virtual assistant;
chatbots that run Skype and answer customer
service queries or deliver info such as weather or
travel updates and the company has rolled out
intelligent features within its Office enterprise.
• Other companies can use the Microsoft AI
Platform to create their own intelligent tools.
• In the future, Microsoft wants to see intelligent
machines with generalized AI capabilities that
allow them to complete any task.

Service - Disney
• Always at the top of delivery extraordinary
service, Disney is getting even better thanks to big
data.
• Every visitor gets their own MagicBand wristband that
serves as ID, hotel room key, tickets, FastPasses and
payment system.
• While guest enough the convenience, Disney gets a lot
of data that helps them anticipate guests’ needs and
deliver an amazing, personalized experience.
• They can resolve traffic jams, give extra services to
guests who may have been inconvenienced by a
closed attraction and data even allows the company to
schedule staff more efficiently.

Social Media - Twitter
• From what tweets to recommend to
fighting inappropriate or racist content
and enhancing the user
experience, Twitter has begun to use
artificial intelligence behind the scenes
to enhance their product.
• They process lots of data through deep
neural networks to learn over time
what users preferences are.

Social Media - Facebook
• Deep learning is helping Facebook
draw value from a larger portion of
its unstructured datasets created by
almost 2 billion people updating
their statuses 293,000 times per
minute.
• Most of its deep learning
technology is built on the Torch
platform that focuses on deep
learning technologies and neural
networks.

Consumer goods - Barbie
• Using natural language processing, machine
learning and advanced analytics, Hello Barbie
listens and responds to a child.
• A microphone on Barbie’s necklace records
what is said and transmits it to the servers at
ToyTalk.
• There, the recording is analyzed to determine
the appropriate response from 8,000 lines of
dialogue.
• Servers transmit the correct response back to
Barbie in under a second so she can respond
to the child.

Consumer Goods – Coca-Cola
• Coca-Cola’s global market and extensive
product list—more than 500 drink brands
sold in more than 200 countries—make it
the largest beverage company in the
world.
• Not only does the company create a lot of
data, it has embraced new technology
and puts that data into practice to
support new product development,
capitalize on artificial intelligence bots
and even trialing augmented reality in
bottling plants.

Consumer Goods - Heineken
• Even though Dutch company Heineken
has been a worldwide brewing leader for
the last 150 years, they are looking to
catapult their success specifically in the
United States by leveraging the vast
amount of data they collect.
• From data-driven marketing to the
Internet of Things to improving
operations through data analytics,
Heineken looks to AI augmentation and
data to improve its operations,
marketing, advertising and customer
service.

Activity
Write one top use case in your
organization

Activity Steps
• Step 1: Group Discussion Industry wise on the use cases
• Step 2: Present the use case with draft implementation plan
• Step 3: Validation of the use case by delegates through Q&A

AI (Data Science) Project Work Flow
65

Data Science Roles
66
Researchers
Data Science Developers Big Data Specialists
Analysts
Business Person
Data Scientist
Infra Engineers

DS Infrastructure Engineers
• Data Science Engineers are hard-core
techies who deal with Data Science
Infrastructure, ie., hardware, software
applications and other aspects to get the
back end of Data Science up and running.
• They set up the entire IT infrastructure from
servers, networks to processes, also manage
it such as infrastructure monitoring,
application management, database
administration etc.,
67

Data Science Developers
68
• Data Science Developers are the ones, who code the models and
applications through programing in R Language, Python etc., They are
versatile developers, who have good knowledge on math & statistics,
machine learning algorithms and related concepts.
• As this domain of data science development is evolving rapidly, these
developers are expected to keep themselves updates with all latest
technological advances from development perspective, so that they can
use the right platform to achieve their goal in effective manner.

Machine Learning Specialist
• These are the professional with deep knowledge
in computer science and mathematics.
• They engage in machine learning and deep
learning heavily.T
• hey create predictive and prescriptive models
based on the machine learning algorithms, such
as random forest, Artificial Neural Network, K-
NN etc.,
• They are masters in all kinds of data mining
techniques pruning, regularization etc., helping
to create a robust data science models, which
can be used in creating great business insights.
69

Big Data Specialists
• Big Data Specialists focus on designing
model for of Big Data Processing as one of
the building block of Data Science work
flow
• Their work involve architecture of Data
gathering including streaming and
snapshots, storing and processing in
effective and efficiently manner using Big
Data Technologies
70

Researchers
• Researchers are domain experts. These
professionals have more expertise in Data
analysis, Data Science related Statistics along
with significant expertise in specific domain
such as HR, Marketing, Fraud Analytics,
Health Care, Finance etc.,
• They have less emphasis and knowledge in
backend part such as IT infrastructure,
coding, computer science etc.,
71

Data Science Analyst
• These professionals are focussed on day-
to-day analysis of data, including website
analytics, retrieving data from various
data sources and creating data
visualizations.
• They work closely with business person.
Their role is to provide the reports from
data analysis with appropriate
visualizations in an easy to understand
format.There by enabling the decision
makers to gain valuable business insights
72

Business Person
• This role is the one, who manages the entire
Data Science project. We can call him as
sponsor. This roles is predominantly
business focussed and, formulates the
problem statements for which, Data science
project needs to find the answers. This role
also helps in understanding, interpreting the
intermediate results of the data science
projects and drive the project to final
solution.
• Though this role is primarily business
focussed, he/she must also speak Data in
order to able to perform the job well.
73

Data Scientist
• These are the professional who is able to
perform every aspect of Data Science,
sometime called as full-stack Data science
professionals. Well, these people are rare as
mastering entire Data Science roles is
difficult, if not impossible.
• If a company manages to hire full-stack Data
Scientist, there would be tremendous
progress in transforming the business in gain
significant competitive advantage by finding
solutions to important business questions.
74

Business
TechnologyAnalytics
Business
Leaders
Delivery
Managers
Visual
Analysts
Process
Consultants
Data
Engineers
Data
Architects
Data
Analysts
Data
Scientists
Lead analytics transformation
across organization
Deliver data- and analytics-driven insights
and interface with end users
Lead analytics transformation
across organization
Build interactive decision-support tools
and implement solutions
Visualize data and build reports
and dashboards
Ensure quality and consistency of
present and future data flows
Collect, structure, and analyse
data
Develop statistical models and algorithms

Economics of Artificial
Intelligence

From Economics of cost perspective - Artificial Intelligence
reduces cost of prediction significantly

Case for Thought : AI - incremental to transformational
• How to identify areas of AI disruption—and value the knob, rather
than turning up the volume, you are instead turning up the
prediction accuracy of the AI.
• Imagine applying the exercise to Amazon’s recommendation engine.
• It is about 5 percent accurate, meaning that out of every 20 things it
recommends, we buy one of them and not the other 19
• Every day people in Amazon’s machine-learning group are working to
crank up that prediction-accuracy knob. You can imagine that knob is
can be tuned up to 30% and 60%.
• We are now sufficiently good at predicting what customer want to
buy. Why are we waiting for you to shop at all? We’ll just ship it.”
• By doing this, Amazon could increase its share of wallet for two
reasons. The first is that it preempts you from buying those goods
from its competitors, either online or offline.
• The second is that, if you were wavering on buying something, now
that it’s on your porch you might think, “Well, I might as well just
keep it.”
AI Disruption
Accuracy :
5% to 30% to 60%

• The single most important question executives in every industry need
to ask themselves is: How fast do I think the knob will turn for a
particularly valuable AI application in my sector?
• If you think it will take 20 years to turn that knob to the
transformational point, then you’ll make a very different set of
investments today than if you think it will take three years.

AI - Limitations and Challenges

First Challenge: Labelling training data
• Labelling data often must be done manually and is necessary for supervised
learning.
• Ironically, machine learning often requires large amounts of manual effort; in
supervised learning, the set of machine learning techniques that is most often
used, machines are taught; they don’t learn “by themselves.”
• Promising new techniques are emerging to address this challenge, such as
reinforcement learning (discussed earlier) and in-stream supervision, in which
data can be labelled in the course of natural usage.

• It is difficult to obtain sufficiently large and comprehensive to be
used for training.
• for many business use cases, creating or obtaining such massive
data sets can be difficult—for example, limited clinical-trial data to
predict health-care treatment outcomes more accurately.
Second Challenge: Requires Large Data Sets

• The difficulty of explaining in human terms results from large and
complex models: why was a certain decision reached?
• Product certifications in health care, as well as in the automotive,
chemicals, and aerospace industries, for example, can be an
obstacle; among other constraints, regulators often want rules and
choice criteria to be clearly explainable.
Third Challenge: Difficulty in explaining results

• AI models continue to have difficulties in carrying their experiences
from one set of circumstances to another.
• That means companies must commit resources to train new models
even for use cases that are similar to previous ones. Transfer
learning—in which an AI model is trained to accomplish a certain
task and then quickly applies that learning to a similar but distinct
activity, is one promising response to this challenge
Fourth Challenge: Generalizability of learning

• while machine-learning algorithms enable companies to realize new
efficiencies, they are as susceptible as any system to the “garbage
in, garbage out” syndrome.
• In the case of self-learning systems, the type of “garbage” is biased
data. Left unchecked, feeding biased data to self-learning systems
can lead to unintended and sometimes dangerous outcomes.
Fifth Challenge: Risk of bias in data and
algorithms

• Societal concerns and regulations can affect potential value capture from AI,
including in use cases related to personally identifiable information.
• This is particularly relevant at a time of growing public debate about the use
and commercialization of individual data on some online platforms.
• Use and storage of personal information is especially sensitive in sectors such
as banking, health care, and pharmaceutical products, as well as in the public
and social sector.
Sixth Challenge: Societal concerns and
regulations.

Pitfalls in AI and Advanced
Analytics Adoption

Pitfall – 1
Top Management lacks vision for AI Initiatives
• This often stems from executives lacking a solid understanding of the difference between
traditional analytics (BI - Reporting) and Advanced Analytics( AI Initiatives)
• To illustrate, one organization had built a centralized capability in advanced analytics, with heavy
investment in data scientists, data engineers, and other key digital roles. The CEO regularly
mentioned that the company was using AI techniques, but never with any specificity.
• In practice, the company ran a lot of pilot AI programs, but not a single one was adopted by the
business at scale.
• REASON: Top management didn’t really grasp the concept of advanced analytics. They struggled
to define valuable problems for the analytics team to solve, and they failed to invest in building
the right skills. As a result, they failed to get traction with their AI pilots. The analytics team they
had assembled wasn’t working on the right problems and wasn’t able to use the latest tools and
techniques. The company halted the initiative after a year as skepticism grew.

1. Whoever is tasked with leading this advanced analytics and AI initiatives—
should set up a series of workshops for the executive team to coach its
members in the key tenets of advanced analytics and to undo any lingering
misconceptions.
2. These workshops can form the foundation of in-house “academies” that can
continually teach key analytics concepts to a broader management audience.
Pitfall – 1: Suggestion
Top Management lacks vision for AI Initiatives

Pitfall – 2
No one has determined the value that the initial use cases can
deliver in the first year
• Too often, the enthusiastic inclination is to apply analytics tools and methods like wallpaper—as
something that hopefully will benefit every corner of the organization to which it is applied.
• But such imprecision leads only to large-scale waste, slower results (if any), and less confidence,
from shareholders and employees alike, that analytics initiatives can add value.
• That was the story at a large conglomerate. The company identified a handful of use cases and
began to put analytics resources against them. But the company did not precisely assess the
feasibility or calculate the business value that these use cases could generate, and, lo and behold,
the ones it chose produced little value.

1. Companies in the early stages of scaling analytics use cases must think through,
in detail, the top three to five feasible use cases that can create the greatest
value quickly—ideally within the first year.
2. This will generate momentum and encourage buy-in for future analytics
investments.
3. These decisions should take into account impact, first and foremost.
4. A helpful way to do this is to analyse the entire value chain of the business,
from supplier to purchase to after-sales service, to pinpoint the highest-value
use cases
Pitfall – 2 : Suggestion
No one has determined the value that the initial use cases can
deliver in the first year

Pitfall – 3
There’s no analytics strategy beyond a few use cases
• In one example, the senior executives of a large manufacturer were excited about advanced
analytics; they had identified several potential cases where they were sure the technology could
add value. However, there was no strategy for how to generate value with analytics beyond those
specific situations.
• Meanwhile, a competitor began using advanced analytics to build a digital platform, partnering
with other manufacturers in a broad ecosystem that enabled entirely new product and service
categories.
• By tackling the company’s analytics opportunities in an unstructured way, the CEO achieved some
returns but missed a chance to capitalize on this much bigger opportunity. Worse yet, the missed
opportunity will now make it much more difficult to energize the company’s workforce to imagine
what transformational opportunities lie ahead.

There are three crucial questions the CDO must ask the company’s business
leaders:
• What threats do technologies such as AI and advanced analytics pose for the
company?
• What are the opportunities to use such technologies to improve existing
businesses?
• How can we use data and analytics to create new opportunities?
There’s no analytics strategy beyond a few use cases

Pitfall – 4
Poorly defined Analytics Roles
• Few executives can describe in detail what analytics talent their organizations have, let alone
where that talent is located, how it’s organized, and whether they have the right skills and titles.
• In one large financial-services firm, the CEO was an enthusiastic supporter of advanced analytics.
He was especially proud that his firm had hired 1,000 data scientists, each at an average loaded
cost of $250,000 a year.
• Later, after it became apparent that the new hires were not delivering what was expected, it was
discovered that they were not, by strict definition, data scientists at all. In practice, 100 true data
scientists, properly assigned in the right roles in the appropriate organization, would have
sufficed.
• Neither the CEO nor the firm’s human-resources group had a clear understanding of the data-
scientist role—nor of other data-centric roles, for that matter

• The right way to approach the talent issue is to think about analytics and AI talent
as a tapestry of skill sets and roles (interactive).
• Naturally, many of these capabilities and roles overlap—some regularly, others
depending on the project.
• Each thread of that tapestry must have its own carefully crafted definition, from
detailed job descriptions to organizational interactions.
• The CDO and chief human resources officer (CHRO) should lead an effort to detail
job descriptions for all the analytics roles needed in the years ahead
Poorly defined Analytics Roles

Pitfall – 5
The organization lacks analytics translators
• Too often, the enthusiastic inclination is to apply analytics tools and methods like wallpaper—as
something that hopefully will benefit every corner of the organization to which it is applied.
• But such imprecision leads only to large-scale waste, slower results (if any), and less confidence,
from shareholders and employees alike, that analytics initiatives can add value.
• That was the story at a large conglomerate. The company identified a handful of use cases and
began to put analytics resources against them. But the company did not precisely assess the
feasibility or calculate the business value that these use cases could generate, and, lo and behold,
the ones it chose produced little value.

1. Hire or train translators right away.
2. Hiring externally might seem like the quickest fix. However, new hires lack the
most important quality of a successful translator: deep company knowledge.
3. The right internal candidates have extensive company knowledge and business
acumen and also the education to understand mathematical models and to
work with data scientists to bring out valuable insights.
4. As this unique combination of skills is hard to find, many companies have
created their own translator academies to train these candidates.
The organization lacks analytics translators

Pitfall – 6
Large scale data-cleansing efforts
• There’s a tendency for business leaders to think that all available data should be scrubbed clean
before analytics initiatives can begin in earnest. Not so.
• Its estimated that companies may be squandering as much as 70 percent of their data-cleansing
efforts. Not long ago, a large organization spent hundreds of millions of dollars and more than
two years on a company-wide data-cleansing and data-lake-development initiative.
• The objective was to have one data meta-model—essentially one source of truth and a common
place for data management. The effort was a waste.
• The firm did not track the data properly and had little sense of which data might work best for
which use cases.
• And even when it had cleansed the data, there were myriad other issues, such as the inability to
fully track the data or understand their context.

1. CDO should orchestrate data cleansing on the data that fuel the most valuable
use cases.
2. In parallel, he or she should work to create an enterprise data ontology and
master data model as use cases become fully operational.
Large scale data-cleansing efforts

Pitfall – 7
No one is hyper focused on identifying potential ethical, social,
and regulatory implications of analytics initiatives
• It is important to be able to anticipate how digital use cases will acquire and consume data and to
understand whether there are any compromises to the regulatory requirements or any ethical
issues.
• One large industrial manufacturer ran afoul of regulators when it developed an algorithm to
predict absenteeism.
• The company meant well; it sought to understand the correlation between job conditions and
absenteeism so it could rethink the work processes that were apt to lead to injuries or illnesses.
• Unfortunately, the algorithms were able to cluster employees based on their ethnicity, region,
and gender, even though such data fields were switched off, and it flagged correlations between
race and absenteeism.

1. As part of a well-run broader risk-management program, the CDO should take
the lead, working with the CHRO and the company’s business-ethics experts
and legal counsel to set up resiliency testing services
2. These services should be tasked to quickly expose and interpret the secondary
effects of the company’s analytics programs.
No one is hyper focused on identifying potential ethical, social,
and regulatory implications of analytics initiatives

AI Data Strategy

Adopt a sophisticated data strategy as data is the
key for any AI based solutions.
• Developing a comprehensive data strategy that focuses not only on
the technology required to pool data from disparate systems but also
on data availability and acquisition, data labeling, and data
governance is the key.
• Although newer techniques promise to reduce the amount of data
required for training AI algorithms, data-hungry supervised learning
remains the most prevalent technique today. And even techniques
that aim to minimize the amount of data required still need some
data.

Data in your organization
• Where does your data live?
• What are the systems governing your data?
• Who is the owner?
• Where are the gaps?

Building Foundations for AI Data
1. Building data lakes
2. Four Stages of Building and Integrating Data Lakes within
Technology Architectures
3. Best Practice Data Architecture

Data Lake
• A data lake is a centralized repository that allows you to store all your
structured and unstructured data at any scale. You can store your
data as-is, without having to first structure the data, and run different
types of analytics—from dashboards and visualizations to big data
processing, real-time analytics, and machine learning to guide better
decisions.

Four Stages of Building
and Integrating
Data Lakes within
Technology
Architectures

Best Practice Reference - Data Architecture

Insights and Closing Remarks

AI - Potential Value Creation
• An analysis of the value derived from McKinsey use cases[1] suggests
that AI can generate considerable value.
• Neural networks and convolutional neural networks—together have
the potential to create between $3.5 trillion and $5.8 trillion in value
annually across nine business functions in 19 industries.
• This constitutes about 40 percent of the overall $9.5 trillion to $15.4
trillion annual impact that could potentially be enabled by all AI and
Advanced Analytics solutions.
[1] Notes from the AI frontier: Applications and value of deep learning. Discussion Paper - McKinsey Global Institute - April 2018

62%
Potential Incremental Value
from AI Over Other analytics
Techniques

Initial AI initiatives are meet
with uncertainties thereby
delaying the projects.

The impact of AI might not be linear but could build up at
an accelerating pace over time.
• AI contribution to growth might be three or
more times higher by 2030 than it is over the
next five years. An S-curve pattern of adoption
and absorption of AI is likely
• a slow start due to the substantial costs and
investment associated with learning and
deploying these technologies, then an
acceleration driven by the cumulative effect of
competition and an improvement in
complementary capabilities alongside process
innovations.
It would be a misjudgement to interpret this “slow burn” pattern of
impact as proof that the effect of AI will be limited.

Transitioning to AI : How big is that challenge?
• In terms of magnitude, it’s akin to coping with the large-scale shift
from agricultural work to manufacturing that occurred in the early
20th century in North America and Europe, and more recently in
China.
• the speed of change today is potentially faster. The task confronting
every economy, will likely be to retrain and redeploy tens of millions
of midcareer, middle-age workers.

How AI could affect countries economies
• Potentially, AI might widen gaps between countries, reinforcing the
current digital divide. Countries might need different strategies and
responses as AI-adoption rates vary.
• Developed economies have higher wage rates, which means that
there is more incentive to substitute labour with machines than there
is in low-wage, developing countries.
• Developed countries could capture an additional 20 to 25 percent in
net economic benefits, compared with today, while developing
countries might capture only about 5 to 15 percent.

Best leaders seize AI Opportunities
• The best leaders, be they visionary or operationally oriented, will
seize this moment to lead their organizations through the most
disruptive period they will experience in their professional lives.
• They will recognize the magnitude of the opportunity, and they will
transform their organizations and industries

Next Practices rather than Best Practices:
It is early, so to talk about best practices might be a little bit
preliminary. We might be talking about next practices, in a
certain sense.

Workplaces and workflows will change as
more people work alongside machines
• As intelligent machines and software are integrated more deeply into
the workplace, workflows and workspaces will continue to evolve to
enable humans and machines to work together.
• As self-checkout machines are introduced in stores, for example,
cashiers can become checkout assistance helpers, who can help
answer questions or troubleshoot the machines

AI and Automation will likely put pressure on
average wages in developed economies
• The occupational mix shifts will likely put pressure on wages. Many of the
current middle-wage jobs in advanced economies are dominated by highly
automatable activities, such as in manufacturing or in accounting, which
are likely to decline.
• High-wage jobs will grow significantly, especially for high-skill medical and
tech or other professionals, but a large portion of jobs expected to be
created, including teachers and nursing aides, typically have lower wage
structures.
• The risk is that automation could exacerbate wage polarization, income
inequality, and the lack of income advancement that has characterized the
past decade across advanced economies, stoking social, and political
tensions.

Jobs without creativity will perish
• It is predicted that most jobs which doesn’t require human
creativity with perish in next 10 to 15 years.

References
• McKinsey DISCUSSION PAPER April 2018
• Michael Chui | San Francisco James Manyika | San Francisco Mehdi Miremadi | Chicago Nicolaus
Henke | London Rita Chung | Silicon Valley Pieter Nel | New York Sankalp Malhotra | New York
• An-executives-guide-to-ai , McKinsey Article, Dec 2018.
• Michael Chui | Vishnu Kamalnath|Brian McCarthy
• https://www.forbes.com/sites/insights-intelai/2018/07/17/from-imitation-games-to-the-real-
thing-a-brief-history-of-machine-learning/#20c22d420563
• ai-business-case-ebook , Published by Gartner.com – 2018
• https://www.forbes.com/sites/nvidia/2019/02/13/how-business-leaders-are-solving-the-ai-
equation
• https://www.forbes.com/sites/bernardmarr/2018/04/30/27-incredible-examples-of-ai-and-
machine-learning-in-practice/#74aad2737502

Q&A?

AI is the next phase of
human evolution
Thank you

A workshop on 'AI transforming Business’

More Related Content

What's hot (20)

Similar to A workshop on 'AI transforming Business’ (20)

Recently uploaded (20)

A workshop on 'AI transforming Business’