View source for Basics-of-Artificial-Intelligence/C2/Machine-Learning-and-Deep-Learning/English

Spoken Tutorial-AI-2

Introduction to Machine Learning and Deep Learning

'''Meta Tags''': AI, Artificial Intelligence, Machine Learning, Deep Learning, Rule-based Programming, Neural Networks, Spam Filter, Voice Assistant, Spoken Tutorial, EduPyramids

'''Pre-requisite Tutorial''' : '''Introduction to Artificial Intelligence'''


{|border=1
|-
|| '''Visual Cue'''
|| '''Narration'''
|-
|| '''Slide 1'''

'''Title slide'''

'''Machine Learning and Deep Learning'''
|| Welcome to this Spoken Tutorial on '''Machine Learning and Deep Learning'''.
|-
|| '''Slide 2 '''

'''Learning Objectives'''

Montage of AI tools from previous tutorial (YouTube, Maps, Photos)
|| In this tutorial, we will learn, 
* About''' Rule-based Programming, Machine Learning''' and '''Deep Learning''', and 
* How '''Machine Learning''' and '''Deep Learning''' are connected to '''AI.'''

|-
|| '''Slide 3'''

'''System Requirements'''

|| For this tutorial, you will need: 
* A computer, a laptop, or a smartphone.
* A stable internet connection.
* An updated web browser, such as Chrome, Edge, or Firefox.

You do '''not''' need coding or technical skills.
|-
|| Image/animation: Show '''Animated gears or code flow'''
|| Let’s start with '''rule-based programming''' using a real-life example.
|-
|| Image/animation: Automatic water dispenser with sensor

|| Think about the time you used an automatic water tap.

You place your hand under a water tap. 

The '''sensor''' in the tap detects your hand. 

Then, it automatically turns on the water tap.

The water starts flowing.
|-
|| Image/animation: Close-up of tap sensor

Zoom in on sensor area; show “Detection = green light
|| The '''sensor''' follows one rule:

“'''IF''' a hand is detected under the tap, '''THEN''' turn on the water tap.”
|-
|| Image/animation: large vessel placed under the tap

Zoom in on sensor area; show “ No Detection”

|| Now, think about what would happen if we placed a large vessel under the tap? 

The sensor in the tap will not detect it. 

This happens because''' rules''' can’t handle new cases.

They have limitations. 

They can only follow the exact instructions we give them.
|-
|| Image/animation: 

Simple flowchart: Hand detected → Tap on → Water flows
|| This is '''rule-based programming'''. 

The machine simply follows the exact set of instructions given to it.
|-
|| Image/animation: : “ Machine Learning” 
|| Now, let’s discuss '''Machine Learning,''' commonly referred to as '''ML'''.
|-
|| Image/animation: Email inbox with spam and regular mails
|| In the '''email inbox''' shown here, do you see a '''folder''' labelled ''''Spam''''?
|-
|| Image/animation: Show Spam folder of an email 

“Data → ML Model → Prediction”

Highlight the suspicious words, links, or senders.

Image/animation:Highlight spam mails vs regular mails (side by side)

Highlight “model” in the diagram
|| Your unwanted''' email '''automatically goes into this folder.

This sorting happens using '''Machine Learning.'''

'''ML''' is trained using thousands of '''spam''' and regular '''emails'''.

From these examples, the machine learns patterns of what spam looks like.

It checks for things like suspicious words, strange links, and unknown senders.

Once trained, it can predict whether a new '''email '''is '''spam '''or not.
|-
|| Pause icon overlay
|| Pause the''' tutorial '''now.

Open the '''Spam''' '''folder''' in your''' email.'''

Look at the '''emails '''in this''' folder'''. 

They have been separated from your''' inbox '''using '''Machine Learning.'''
|-
|| Transition slide:Deep Learning
|| Now, let’s go one step deeper into '''Machine Learning''' and understand '''Deep Learning'''.

'''Deep Learning''' is commonly referred to as '''DL.'''
|-
|| Visual: 
* Voice assistant (e.g. Google Assistant / Siri) 
* what is the time in India now
* Show the response

|| Let’s open '''Google Chrome'''. 

Click on the '''mic icon '''in the''' search bar''' to ask a question in''' voice mode. '''

Then ask, “What’s the time in India now?”

It understands and processes the voice input and gives an answer.

This process happens with the help of '''Deep Learning'''.
|-
|| '''Visual:''' Illustration of neural network (dots and lines)

Animation: 
* Small dots connected by lines
* Small dots lighting up like decision nodes
* Arrows between nodes passing signals
* Layers lighting up one after another
* “Many Layers = Deep Learning”

|| Do you know how deep learning gave the answer? 

'''Deep Learning''' uses '''artificial neural networks. '''

These networks have many layers that work together.

You can think of them as many tiny helpers.

Each helper understands a small part of the voice.

They pass this information from one layer to the next.

As information travels through the layers, the understanding becomes clearer.

Finally, the computer understands your question and gives the right answer.
|-
|| Visual
* Layers lighting up one after another
* “Many Layers = Deep Learning”
* Show recognizing faces or understanding speech by computer

|| As there are many layers, we refer to it as "'''deep."'''

Each layer helps improve the understanding a little more, like building blocks.

Hence, '''Deep Learning '''is used for complex tasks.

For example, in understanding speech or recognizing images.
|-
|| Visual: Russian doll animation: AI (outer) → ML (inner) → DL (innermost)
|| Imagine '''AI, ML''', and '''DL''' like Russian dolls.

One doll inside the other. 

First comes '''AI''', the biggest doll.

Inside '''AI''' is '''ML''', and inside ML is '''DL.'''
|-
|| Visual: Highlight outer doll: AI
|| '''AI''' is a broad field that makes machines intelligent. 

It’s like the outer shell that includes many different methods.
|-
|| Visual: Highlight middle doll: ML
|| '''ML''' is a '''subset '''of''' AI.'''

It learns from data. 

It’s a smaller doll inside '''AI'''.
|-
|| Visual: Highlight inner doll: DL
|| '''Deep Learning''' is a '''subset '''of '''ML'''. 

It is nested inside '''ML'''.

It learns from large amounts of data using artificial neural networks.

This lets it handle complex tasks.
|-
|| '''Summary slide:'''

'''Bulleted List '''
|| Let’s quickly summarize what we learned in this tutorial.

We learned that:
* '''Rule-based programming''' follows fixed steps.
* '''Machine Learning''' learns patterns from data.
* '''Deep Learning''' uses neural networks with many layers.
* '''ML''' and '''DL''' are parts of '''AI'''

|-
|| '''Assignment slide'''
* Think of an application from your daily life that uses '''fixed rules'''. (This is your rule-based programming example.)
* Next, think of another application that '''learns from data''' or uses '''Deep Learning'''. (Examples include spam filters, image recognition, or voice assistants.)
* Write down both examples and explain why each one fits its category.

|| Now, here’s a quick assignment for you.

Think of an application from your daily life that uses '''fixed rules.'''

Next, think of an application that learns from data or uses deep learning. 

Write down both examples and explain why each one fits its category.
|-
|| EduPyramids logo
|| This Spoken Tutorial is brought to you by '''EduPyramids Educational Services Private Limited'''.
|-
|| Closing slide
|| Thank you for watching!
|-
|}