6.

5 Artificial intelligence &

machine learning
 ARTIFFICIAL
INTELLIGENCE

WhatAIISis artificial
Definition: a branch of intelligence
computer science aimed at building
intelligent machines that simulate human behavior and thought
processes, capable of making decisions autonomously.

Artificial Intelligence is composed of two words Artificial and

Intelligence, where Artificial defines "man-made," and intelligence
defines "thinking power", hence AI means "a man-made thinking
power.“

AI holds a tendency to cause a machine to work as a human

With AI you do not need to pre-program a machine to do some

work, despite that you can create a machine with programmed
algorithms which can work with own intelligence and that is the
awesomeness of AI
 Artificial learning
IMPORTANCE GOAL COMPRISES OF
• create such software or • Replicate human • not just a part of
devices which can solve intelligence computer science even
real-world problems • Solve Knowledge- it's so vast and requires
very easily and with intensive tasks lots of other factors
accuracy such as health • An intelligent which can contribute to
issues, marketing, connection of it
traffic issues, perception and action • intangible part of our
• you can build such • perform tasks that brain which is a
robots which can work requires human combination of
in an environment intelligence reasoning, learning,
where survival of • Creating some system problem-solving
humans can be at risk which can exhibit perception, language
• you can create your intelligent behaviour, understanding, etc.
personal virtual learn new things by
assistant, such as itself, demonstrate,
Cortana, Google explain, and can advise
Assistant, Siri, etc. 3
to its user
Advantages and disadvantages of artifficial
learning (AI)

ADVANTAGES DISADVANTAGES

High accuracy with less

High cost
error

High -speed Cant think out of box

High reliability No feelings and emotions

Increase dependency on
Useful for risky areas
machines

Digital assistant No original creativity 4

Differences: AI VS Machine
Learning
ARTIFICIAL INTELLIGENCE MACHINE LEARNING
What is it? AI is broad term for machine- ML is an artificial
based applications that mimic intelligence methodology.
human intelligence. Not all AI All ML solutions are AI
solutions are ML. solutions.
Best suited for? AI is best for completing a ML is best for identifying
complex human task with patterns in large sets of
efficiency. data to solve specific
problems.
Methods AI may use a wide range of For ML, people manually
methods, like rule-based, neural select and extract features
networks, computer vision, and from raw data and assign
so on. weights to train the model.
Implementation AI implementation depends on You train new or existing
the task. AI is often prebuilt and ML models for your specific
accessed via APIs. use case. Prebuilt ML APIs
are available.
 Machine Learning
- Is an approach to data analysis that involves
building and adapting models, which allows
programs to 'learn' through experience.
- Involves the construction of algorithms that
adapt their models to improve their ability to
make predictions.

How does Machine Learning

• Machine Learning models like Deep Learning allow works??
the vast
majority of data to be handled with an accurate generation of
predictions. ​
• These Machine Learning algorithms use the patterns contained
in the training data to perform classification and future
predictions. ​
• Whenever any new input is introduced to the Machine Leaning
model, it applies its learned patterns over the new data to make
future predictions. ​
• Based on the final accuracy, one can optimize their models
Machine Learning &
Algorithm
Is classified to three
types:

Supervised Unsupervised
Learning Learning

Reinforcement
Learning

8
Supervised Learning Unsupervised Learning

- An example of - there is no labeled data.

supervised learning is - The algorithm identifies the patterns within the
spam filtering. dataset and learns them.
- This is an easier form of - The algorithm groups the data into various clusters
training. Spam filtering based on their density.
is an example of this -
type of machine Using it, one can perform visualization on high
dimensional data.
learning algorithm.
- One example of this type of Machine learning
algorithm is the Principle Component Analysis.
- Furthermore, K-Means Clustering is another type of
Unsupervised Learning where the data is clustered in
groups of a similar order.
- The learning process in Unsupervised Learning is
solely on the basis of finding patterns in the data.
- After learning the patterns, the model then makes
conclusions.
 - Is a type of machine

Reinforcement learning where an agent

(robot) learns how to make
decisions by interacting with

Learning -
the environment
The robot needs to learn by
itself which way is the best
by interacting with the
environment

Game Playing Self-driving Robotics Finance

cars (Trading
Algorithms)
 Classification
Supervised
Learning
MACHINE Regression
LEARNING
Unsupervise
Clustering
d Learning

11
Classification Regression Clustering
Support Vector K-Means
Linear Regression, GLM
Machines
Hierarchical
Discriminant Analysis SVR, GPR
Gaussion Mixture
Naïve Bayes Ensemble Mthods
Hidden Markov Model
Nearest Neighbor Decision Trees
Neural Networks

Neural Networks Neural Networks

12
 Applications of Machine Learning

Traffic Prediction

Search Engine Banking & Finance

Medical Diagnosis Image Recognition

Virtual Personal
Assistant
Real World Application of Machine Learning

14
 Examples of Application

Cyber Alexa by
Waze
Security Amazon

15
Advantages & Disadvantages of Machine
Learning

ADVANTAGES DISADVANTAGES

Identification of trends
Data acquisition
and patterns

No human interference is
Time and resources
required

Continuous improvement Interpretation

Wide application

16
 Thank You!
"What all of us have to do is
to make sure we are using AI
in a way that is for the
benefit of humanity ,Not to
detriment of humanity"

PRESENTED BY:
ADDAH
-Tim Cook
NURUL
MAYRAA