Module 3- AI and ML

Learning Objectives
Concepts Covered:
• Artificial Intelligence (AI)
• Business Applications of AI
• Open AI
• Machine Learning (ML)
• Natural Language Processing (NLP)
• Deep Learning
• AI Techniques:
• Neural Networks
• Fuzzy Logic
• Expert Systems
Introduction to AI
• Definition: "The science and engineering of making intelligent machines,
especially intelligent computer programs." - John McCarthy
• Purpose: AI aims to understand intelligent entities and apply this
understanding to create intelligent machines.
• Fields Involved:
• Perception
• Logical Reasoning
• Specific tasks (e.g., chess, theorem proving, poetry, diagnostics)
“AI is basically the study of training your machine(computers) to
mimic a human brain and its thinking capabilities”.
History of AI
• Origins: Dartmouth Summer Research Project on AI (1950s)
• Key Figures: Alan Turing, Allen Newell, Herbert A. Simon
• Milestone: IBM's Deep Blue defeated chess champion Garry Kasparov
(1996)
• Current Trends: AI in consumer electronics and AI chips
Definition and Meaning of AI
• Coined by: John McCarthy (1956)
• Definition: AI emphasizes creating intelligent machines that work and
react like humans.
• Key Activities:
• Speech Recognition
• Learning
• Planning
• Problem Solving
Turing Test
• Concept: A machine's ability to exhibit intelligent behavior
indistinguishable from a human.
• Procedure: A human judge converses with both a human and a
machine through text. If the judge cannot reliably distinguish
between them, the machine is considered intelligent.
AI in Daily Life
• Examples:
• Virtual Personal Assistants (e.g., Amazon Alexa)
• Video Games
• Smart Cars (e.g., Google driverless car)
• Online Customer Support (e.g., chatbots)
• News Generation (e.g., Facebook news feed)
• Music and Movie Recommendations
• Smart Home Devices
 AI in Action: Siri and DeepMind
Siri: The Power of Natural Language DeepMind: Learning from Experience
Processing- Google's DeepMind utilizes a sophisticated form of
Apple's virtual assistant, Siri, utilizes natural machine learning called deep learning. Deep learning
language processing (NLP) to understand your voice algorithms can learn complex patterns from large
commands. NLP allows Siri to interpret your words, datasets without explicit programming. In essence,
analyze their meaning, and respond in a way that DeepMind can "think" like humans, drawing
feels natural and conversational. This technology connections and gaining insights from raw data. This
powers countless interactions with Siri, enabling you technology has enabled Google to optimize its data
to set reminders, make calls, and access information centers, reducing energy consumption by a significant
through voice commands margin.
Business Application of AI
Artificial intelligence (AI) is no longer a futuristic concept, it's a powerful
force shaping the business world. From automating tasks to unlocking
insights, AI is transforming industries and creating new possibilities. But
what exactly is AI, and how does it work? AI encompasses a wide range of
technologies, including machine learning, deep learning, robotics, and
natural language processing.
These technologies enable machines to learn from data, make decisions,
and even perform tasks that traditionally required human intelligence.
This presentation explores the exciting world of AI and its diverse
applications across various business sectors.
1) AI in Banking and Finance
Many banks use the various applications of artificial intelligence to detect
fraudulent activity. The AI software is given a very large sample of data that
includes fraudulent and non-fraudulent purchases and is trained to
determine whether a transaction is valid based on data. Over time, the
software becomes incredibly adept at spotting fraudulent transactions
based on what it has learned previously, invest in stocks, and manage
properties.
Few example of AI used in bank are:
• detect charges or claims outside of the norm
• AML Pattern Detection
• Chat bots
• Customer recommendations
2) AI in Defense and Security
As cyber-attacks increase in frequency and more sophisticated tools are
used to breach cyber defense, human operators are no longer enough. Top
firms across the world are investing heavily in cybersecurity to ensure their
data is protected. Real-time threat detection, mitigation, and ideally,
prevention, are what’s needed for businesses – and AI can deliver.
Using machine learning algorithms and feeding those algorithms great
quantities of data, IT and security experts can teach the AI solution to
monitor behaviour, detect anomalies, adapt and respond to threats and
issue alerts. AI has quickly become a key component in a business’
cybersecurity infrastructure, providing a multi- layered security strategy
that is robust and sophisticated. Artificial intelligence is now an essential
component of defense and security strategies — from combat systems to
operational processes.
1)The United States Department of Defense (DoD) has pioneered innovative uses for AI
in defense and security, When it comes to analyzing video and images captured by
surveillance systems and aerial vehicles, AI is a huge advantage. For example,
algorithms can be trained to recognize terrorist activity evident in streams of video, just
as they can be trained to recognize cats in datasets filled with all kinds of images.
2) AI can help military organizations combat the threat of cyber-attacks, which can now
be launched from virtually anywhere in the world.
3) The U.S. Department of Homeland Security has piloted AI tools for detecting cyber-
network intrusions and malicious activities.
4) Predictive maintenance. In the recruiting process, the U.S. Army uses an AI-driven
interactive virtual assistant to answer questions from prospective recruits, check users’
qualifications, and refer prospects to human recruiters.
5) In a report on AI-augmented government, Deloitte says that the interactive virtual
assistant does the work of 55 recruiters, with an accuracy rate of more than 94
percent. In military operations, reconnaissance or scouting is the exploration outside
an area occupied by friendly forces to gain information about natural features and
other activities in the area.
3) AI in Hospitals and medicine
Artificial intelligence (AI) in healthcare is the use of algorithms and
software to approximate human cognition in the analysis of complex
medical data.
The primary aim of health-related AI applications is to analyze
relationships between prevention or treatment techniques and patient
outcomes.
a. To organize bed schedules, make a staff rotation, and provide medical
information.
b. Artificial neural networks are used as clinical decision support
systems for medical diagnosis.
c. Detection of a tumor
d. Heart sound analysis.
4) AI in Heavy industry
To create mechanism that is not subject to human feeling like fatigue
and worry. Routine jobs that create mental fatigue may be performed
by Artificial intelligence systems like robots. Robots have proven
effective in jobs that are very repetitive which may lead to mistakes or
accidents due to a lapse in concentration and other jobs which humans
may find degrading. Japan is the leader in using and producing robots
in the world. In 1999, 1,700,000 robots were in use worldwide.
Robotics has already changed the face of manufacturing.
However, robots are becoming far more intelligent and autonomous,
thanks to AI. What is machine learning used for in factories? Many
companies are building so-called “smart manufacturing” facilities that
use AI to optimize labor, speed production and improve
For example, Siemens is now equipping gas turbine systems with more
than 500 sensors that continuously monitor devices and machines. All this
data is helping create the manufacturing facility of the future, sometimes
referred to as Industry 4.0.
Smart manufacturing–which merges the industrial IoT and AI–is projected
to grow from $200 billion in 2018 to $320 billion by 2020, according to a
study conducted by market research firm TrendForce.
5) AI in sports
The applications of artificial
intelligence in sports have become a
common sight in recent years.
Considering the positive impact
they’ve brought about through their
growing capabilities, they will
continue to make inroads into the
realm of sports. Following are a few
areas in sports where artificial
intelligence is set to become a
mainstay component:
6) AI in Marketing
The use of real-time data, Web data, historical purchase data, app use
data, unstructured data and geolocation information have introduced the
ability to deliver information, product recommendations, coupons and
incentives at the right time and place. AI allows companies to engage in
personalized marketing and slide the dial closer to one-to-one
1. Product/content recommendations :Fast forward to 2019, and some of the
most successful digital companies have built their product offerings around the
ability to provide highly relevant and personalised product or content
recommendations – including Amazon, Netflix and Spotify.This all comes from
AI-based clustering and interpreting of consumer data paired with profile
information and demographics.
2. Dynamic pricing :Dynamic pricing (often referred to as personalized pricing)
is a pricing strategy where the price is determined depending upon the
demand, the availability and the profile of the customer. AI technologies are
even used to decide and design personalized offers based on customer profiles.
3. AI-enhanced PPC (pay –per –click) advertising: AI-powered systems can
help advertisers test out more ad platforms and optimize targeting. That’s
exactly what Facebook is doing with their ad delivery optimization. However,
this approach could also be applied to omnichannel PPC campaign data (held
by a single company) by using third-party or in-house AI tools.
7) AI in Retail Sales
Artificial intelligence in retail is transforming the way people shop and buy items ranging from
clothes to cars. Voice search and image search are now widespread. Amazon and many other
retailers now incorporate these tools in their apps. Next generation AI is also taking shape.
For example, augmented reality (AR) lets shoppers view a sofa or paint color superimposed in
their house or office.
Virtual reality (VR) allows consumers to sit inside a vehicle and even test drive it without
leaving home.
Audi, BMW and others have developed VR systems for shoppers. But the AI use cases don’t
stop there.
AI in retail extends to bots and virtual assistants that recommend products and provide
information; algorithms that helps sales teams focus on high value customers and high
probability transactions; and predictive analytics that factor in weather, the price of raw goods
and components, or inventory levels to adjust pricing and promotions dynamically.
Clothing retailer North Face, for instance, asks customers a series of questions related to a
purchase at its website. Not only does this lead customers to the right product, it taps
machine learning to gain insights that potentially lead to higher cart values and additional
sales.
8) AI for Customer Support
AI in retail is emerging as a powerful force, but customer
support is also harnessing the technology for competitive
advantage. Bots and digital assistants are transforming the
way support functions take place. These technologies
increasingly rely on natural language processing to identify
problems and engage in automated conversations.
AI algorithms determine how to direct the conversation or
route the call to the right human agent, who has the
required information on hand. This helps shorten calls and it
produces higher customer satisfaction rates.
A Forrester study found that 73 percent of customers said
that valuing their time is the most important thing a
company can do to provide them with good online customer
service.
9) Artificial intelligence in eCommerce
1. Customer-centric visual search Consumers are often disappointed with
e-commerce experience because the product results showed are often
irrelevant. To tackle this problem, AI uses natural language processing to
narrow, contextualize and improve search results for online shoppers.
Also, it allows having visual search capabilities, finding and matching
products. For example Amazon has this option, which allows you to
point at product that you like and Amazon will identify it and gives
you the results you will most probably like because it will be exactly
what you were searching for.
2. Retarget potential customers and improve the sales process
This means that pre-qualified potential buyers interested in a product
are just left behind. Also, many teams are overloaded with
unmanageable customer data that they do little or nothing with. And
that’s when AI is a real necessity.
Artificial intelligence could help with enhancing the sales cycle, by
tailoring your problem-solving solutions and creating a strong sales
message that reaches consumers at the right time on the right platform.
Example – The North Face, a large e-commerce retailer. They use IBM`s
AI solution Watson to better understand their consumers. IBM’s
software then scans hundreds of products to find perfect matches based
on real-time customer input and does additional research to find out the
weather conditions in the that area, etc.
3.New level of personalization
Now we can find a lot of AI
solutions, for example one of the
them that I really want to mention
is Boomtrain. This company
analyzes different touch points to
help the business analyze how
customers are interacting online:
via mobile app, web, email, etc.
4. Chatbots and virtual assistants
E-Commerce is now focused on user experience. The use
of artificial intelligence through the application of
“chatbots” is just one way to drive the conversation in the
era of conversational commerce. And it is actually more
than that, chatbots can automate order processes.
Also, they are an effective and low-cost way of providing
24/7 customer service and collect valuable data and track
behavior. With chatbot, e- commerce sites can increase
conversion rates by tailoring the online experience for the
consumer. Examples is, Amazon’s virtual assistant, Alexa.
It has integrated into Amazon’s own products as well as
products from other manufacturers. Virtual assistants are
expected to impact customers purchase and provide a
creative opportunity for e-commerce retailers to
10) Improve recommendations for customers
With AI brands can more intelligently and efficiently predict customer behavior
and demand, and offer relevant and helpful recommendations. Example is
Starbucks, it uses AI to analyze all the data it has gathered to deliver more
personalized suggestions.
The algorithm takes into account: customer’s information, customer
preferences, purchase history, third-party data and contextual information.
Filter fake reviews
Unfortunately, fake reviews became an issue for online retailers and e-
commerce shops. However, as we know there are fake reviews published
by competitors, bots, etc. So, how AI can manage this problem?
Nowadays, many e-commerce business use artificial intelligence to fight
astroturfing by putting more emphasis on verified and helpful reviews.
For example, Amazon uses AI to combat fake product reviews and
inflation of their popular star ratings. Their AI focuses on prominence and
weight of verified customer purchase reviews and boost them.
11) Automation
AI doesn’t mean that the robots are taking over. Nowadays a lot of people
are scared that everything will be done by robots in the nearest future.
Robots offer opportunity to retailers to provide exactly what the customer
wants, when they want it, by using technology and algorithms.
As firms in the e-commerce industry continue to expand and that means
having a growing number of customers, automation becomes a must and
became one of their top investment priorities for e-commerce stores.
12) AI in Judiciary
AI is increasingly being utilized in judicial systems around the world
to enhance efficiency and accuracy. Examples include:
1.Supreme Court Vidhik Anuvaad Software (SUVAS) - An AI
system in India that assists in translating legal judgments into
regional languages.
2.China’s Smart Court system - Utilizes AI to analyze past cases
and suggest relevant laws and precedents to judges, aiding in
decision-making processes.
3.AI tools in various judicial bodies - Many jurisdictions are
exploring AI for tasks such as legal research, predicting case
outcomes, and automating administrative processes.
4. SUPACE (Supreme Court Portal for Assistance in Court's
Efficiency)
It's important to double-check specific implementations and their
effects in different legal systems.
13) AI in automobile
14) AI in Education

Analyzing large amounts of data quickly and accurately

Enhancing efficiency of administrative tasks

Facilitating adaptive learning

Providing immediate feedback and support

Machine Learning:
• Machine Learning is basically the study/process which provides the
system(computer) to learn automatically on its own through experiences it had and
improve accordingly without being explicitly programmed.
• ML is an application or subset of AI.
ML focuses on the development of programs so that it can access data to use it for itself. The
entire process makes observations on data to identify the possible patterns being formed and
make better future decisions as per the examples provided to them. The major aim of ML is to
allow the systems to learn by themselves through experience without any kind of human
intervention or assistance.
• With the help of sample historical data, which is known as training data, machine
learning algorithms build a mathematical model that helps in making predictions or
decisions without being explicitly programmed.
• Machine learning brings computer science and statistics together for creating
predictive models. Machine learning constructs or uses the algorithms that learn
from historical data. The more we will provide the information, the higher will be the
performance.
How does Machine Learning work:
As machine learning system receives new data, it forecasts the outcome
using the prediction models it has built using prior data.
1. Input Data: The process begins with collecting raw data relevant to
the problem you're trying to solve.
2. Feature Extraction: Important features or characteristics are extracted
from the raw data to help the algorithm understand the problem better.
3. Model Building: Using the extracted features, the machine learning
algorithm builds a model by learning patterns from the data.
4. Training: The model is trained using a training dataset, which involves
adjusting the model's parameters to improve its accuracy.
5. Testing/Validation: After training, the model is tested with a separate
dataset to evaluate its performance and accuracy.
6. Prediction: Once the model is trained and validated, it can be used to
predict outcomes based on new, unseen data.
7. Output: The final prediction or result is produced, which can be
interpreted and used for decision-making.
• In essence, the diagram shows how data is processed through various
stages to build a machine learning model that can make accurate
predictions.
Problem: Predicting House Prices
1. Input Data
• Suppose you have a dataset with various features related to houses, such as square footage, number of bedrooms, location, age of the
house, etc.
2. Feature Extraction
• From the dataset, you extract key features like:
• Square footage
• Number of bedrooms
• Location (e.g., urban, suburban, rural)
• Age of the house
3. Model Building
• You choose a machine learning algorithm, such as Linear Regression, to model the relationship between these features and the house price.
4. Training
• The algorithm is trained using historical data, where the features (square footage, number of bedrooms, etc.) are used to predict the known
house prices. The model learns the relationship between the features and the price.
5. Testing/Validation
• After training, the model is tested with a different set of data (not used in training) to check how well it predicts house prices. This step helps
in tweaking the model if necessary.
6. Prediction
• Now, suppose you have a new house with known features (e.g., 2000 square feet, 3 bedrooms, suburban area, 10 years old). You input these
features into the trained model.
7. Output
• The model predicts the price of this house based on the learned relationships. For example, it might predict that the house would sell for
$300,000.
Summary:
• By feeding the model with enough data on various houses and their sale prices, the algorithm can learn to predict the price of a new house
1. Supervised learning
A model is trained on a labelled data set using supervised learning, where
the right result is delivered for each input. With the use of this knowledge,
the algorithm can figure out how inputs and outputs are related and make
predictions on previously unobserved data. The system builds a model
using labelled data to comprehend the datasets and learn about each one.
After training and processing, the model is tested by utilising sample data
to see if it accurately predicts the desired outcome. In supervised learning,
mapping input and output data is the main objective. The foundation of
supervised learning is supervision, just like when a student is studying
under a teacher's supervision. Spam filtering is a prime example of
supervised learning.
Other examples: Image and speech recognition,
recommendation systems, and fraud detection
Supervised learning
2. Unsupervised learning:
A model entails building a model on a set of unlabeled data without
knowing which output is intended. The algorithms used for clustering,
dimensionality reduction, and anomaly detection must independently
discover the structure in the data. The machine is trained using a set of
unlabeled, unclassified, or uncategorized data, and the algorithm is
required to respond independently on that data. Unsupervised
learning's objective is to reorganise the input data into fresh features or
a collection of objects with related patterns.
3. Reinforcement Learning:
A learning agent in a reinforcement learning system receives a reward
for each correct action and receives a penalty for each incorrect
activity. With the help of these feedbacks, the agent automatically
learns and performs better. The agent explores and engages with the
environment during reinforcement learning. An agent performs better
since its objective is to accrue the most reward points.
Deep Learning:
• Deep Learning is basically a sub-part
of the broader family of Machine
Learning which makes use of Neural
Networks(similar to the neurons
working in our brain) to mimic
human brain-like behavior.
• DL algorithms focus on information
processing patterns mechanism to
possibly identify the patterns just
like our human brain does and
classifies the information
accordingly.
• DL works on larger sets of data when
compared to ML and the prediction
mechanism is self-administered by
machines.
Applications of Deep Learning
NATURAL LANGUAGE PROCESSING (NLP): Humans
communicate with each other using words and text. The
way that humans convey information to each other is
called Natural Language. Every day humans share a large
quality of information with each other in various
languages as speech or text. However, computers
cannot interpret this data, which is in natural language,
as they communicate in 1s and 0s.

The data produced is precious and can offer valuable

insights. Hence, you need computers to be able to
understand, emulate and respond intelligently to human
speech. Natural Language Processing or NLP refers to
the branch of Artificial Intelligence that gives the
machines the ability to read, understand and derive
meaning from human languages. NLP combines the field
of linguistics and computer science to decipher language
structure and guidelines and to make models which can
comprehend, break down and separate significant
details from text and speech
• In general terms, NLP is a sub-discipline of AI which
focuses on the ability of computers to analyze,
understand, and generate natural language.
• It helps machines interact with humans in a way that
they both can understand – by interpreting the words
spoken by us or written in texts etc.
• In simpler terms, NLP enables machines to read our
queries and respond accordingly in a meaningful way.
• The importance of NLP lies in its potential applications
like providing virtual assistants for customer service
inquiries or providing accurate translations from one
language to another in real-time.
1. Question Answering
Question Answering focuses on building systems that automatically answer the
questions asked by humans in a natural language.
2. Spam Detection
Spam detection is used to detect unwanted e-mails getting to a user's inbox.
3. Sentiment Analysis
Sentiment Analysis is also known as opinion mining. It is used on the web to
analyze the attitude, behaviour, and emotional state of the sender. This application
is implemented through a combination of NLP (Natural Language Processing) and
statistics by assigning the values to the text (positive, negative, or natural), identify
the mood of the context (happy, sad, angry, etc.)
4. Machine Translation
Machine translation is used to translate text or speech from one natural language
to another natural language. Example: Google Translator
5. Spelling correction
Microsoft Corporation provides word processor software like MS-word, PowerPoint .
6. Speech Recognition
Speech recognition is used for converting spoken words into text. It is used in applications,
such as mobile, home automation, video recovery, dictating to Microsoft Word, voice
biometrics, voice user interface, and so on.
7. Chatbot
Implementing the Chatbot is one of the important applications of NLP. It is used by many
companies to provide the customer's chat services.
8. Information extraction
Information extraction is one of the most important applications of NLP. It is used for
extracting structured information from unstructured or semi-structured machine-readable
documents.
9. Natural Language Understanding (NLU)
It converts a large set of text into more formal representations such as first-order logic
structures that are easier for the computer programs to manipulate notations of the
natural
ARTIFICIAL INTELLIGENCE TECHNIQUES
There are various techniques of Artificial intelligence which are classified
into three group
1. Expert systems
2. Neural Network
3. Fuzzy logic
1. Expert Systems:
The classic AI problem of programming has been very successfully
handled by the area of expert system. Experts are those people who
have expertise in certain area possessing knowledge of special skills in
solving various kinds of problems more efficiently and cheaply. In 1970s,
expert systems were in research areas which were slowly being
developed containing expert knowledge exclusively for their domain.
The term expert system is, however, often applied today to any systems
that use expert system technology.
The expert system technology may include special system languages,
programs, and hardware designed to assistance in the development and
execution of expert system. Unlike the general purpose problem solver,
an expert system functions well in their restricted domains for which
they designed for.
User Interface
User interface provides interaction between user of the ES and the ES itself. It is generally
Natural Language Processing so as to be used by the user who is well-versed in the task
domain. The user of the ES need not be necessarily an expert in Artificial Intelligence. It
explains how the ES has arrived at a particular recommendation. The explanation may
appear in the following forms − Natural language displayed on screen.
Interface Engine
Use of efficient procedures and rules by the Interface Engine is essential in deducting a
correct, flawless solution. In case of knowledge-based ES, the Interface Engine acquires
and manipulates the knowledge from the knowledge base to arrive at a particular
solution. In case of rule based ES, it − Applies rules repeatedly to the facts, which are
obtained from earlier rule application.
Knowledge Base
It contains domain-specific and high-quality knowledge. Knowledge is required to exhibit
intelligence. The success of any ES majorly depends upon the collection of highly
accurate and precise knowledge
Components of Knowledge Base
The knowledge base of an ES is a store of both, factual and heuristic
knowledge.
Factual Knowledge − It is the information widely accepted by the
Knowledge Engineers and scholars in the task domain.
Heuristic Knowledge − It is about practice, accurate judgement, one’s
ability of evaluation, and guessing.
Applications:
▪ Introduction to Machine Learning Approaches
▪ Fuzzy Logic Improves Decision Support Software
▪ Shell Programming in Expert Systems Applications
▪ Smart Home Appliances for Better Quality of Life – Combining artificial
intelligence with home automation in smart home appliances results in an
improved quality of life for many, including the elderly and disabled.
▪ Voice Recognition Software for Disabled Students – Disabled students are
often at a disadvantage in the classroom. Voice recognition software
improves communication, enables note-taking, and increases participation.
▪ Teaching Special Needs Children with Autism – Robots are acting as
therapy assistants to help parents and therapists in teaching special needs
children with autism
2. NEURAL NETWORK
The inventor of the first neurocomputer, Dr. Robert Hecht-Nielsen, defines
a neural network as − "...a computing system made up of a number of
simple, highly interconnected processing elements, which process
information by their dynamic state response to external inputs.”
Neural network stimulate human ability to classify things and capable of
finding and differentiating patterns. It is widely used for visual pattern and
speech recognition systems. They are called neural network because the
resemble the structure and functioning of human brain