Introduction

to Machine
Learning
 Why Machine Learning Matters
1. Solves Complex Problems Automatically
ML can solve problems that are too complex for humans to manually code solutions for—like recognizing
faces, predicting disease from symptoms, or detecting spam emails.

2. Data-Driven Decisions
With massive amounts of data generated every second (IoT, apps, sensors, transactions), ML helps analyze and
convert data into actionable insights.

3. Automation & Efficiency

Manual tasks like document scanning, quality checks, or customer queries can be automated with ML.
Reduces human error and saves time and cost.
 Industry Adoption
ML is powering industries like:

• Healthcare: Disease diagnosis, drug discovery

• Retail: Inventory prediction, demand forecasting

• Transportation: Self-driving cars, route optimization

• Agriculture: Crop monitoring, weather predictions

 Data Scientists are focused on
Data Analysts are more
developing and implementing
concerned with interpreting and
predictive models using
presenting data to answer
advanced statistical and
specific business questions and
machine learning techniques to
guide decision-making based on
extract insights and make
historical information.
predictions for future scenarios.
 Supervised Learning

Types of
Machine Unsupervised Learning

Learning
Reinforcenent Learning
Machine Learning is the science (and art) of programming computers so
they can learn from data.

Here is a slightly more general definition:

[Machine Learning is the] field of study that gives computers the ability to
learn without being explicitly programmed.
—Arthur Sanuel, 1G5G
 How Machine Learning Works (High-Level)

1. Data Collection: Gather raw information (e.g., student scores, customer transactions)
2. Preprocessing: Clean and structure the data
3. Model Selection: Choose an algorithm (e.g., linear regression, decision tree)
4. Training: Feed the algorithm with training data
5. Evaluation: Test model accuracy on new/unseen data
6. Prediction or Decision Making: Use the model to predict future outcomes.
Imagine you have a friend who loves recommending movies to
you. At first, you tell your friend about the types of movies you
enjoy, and they start suggesting films based on that information.
Over time, as you watch more movies and share your opinions,
your friend starts to understand your preferences better. They
learn from the movies you liked and didn't like.

Now, instead of you explicitly telling your friend what you want to
watch each time, your friend has become so good at
understanding your taste that they can make suggestions without
you providing specific instructions. They've learned to predict
what you might enjoy based on patterns in your past preferences.

In this analogy:

• You: Represent the computer or the system that is learning.

• Your friend: Represents the machine learning model.
• Movies and your preferences: Are the data used for learning.
Supervised Learning:

Definition: In supervised learning, the algorithm is

trained on a labeled dataset, where the input data is
paired with corresponding output labels. The goal is
for the algorithm to learn the mapping from inputs to
outputs so that it can make accurate predictions on
new, unseen data.
Exanple:
• Use Case: Email Spam Classification
• Explanation: Given a dataset of emails labeled as
"spam" or "not spam," a supervised learning
algorithm can learn to classify new, unseen emails
into these categories based on features such as
words, sender information, and email structure.
Unsupervised Learning:
Definition: Unsupervised learning involves training an
algorithm on unlabeled data, and the system tries to find
patterns, relationships, or structures within the data without
explicit guidance on what to look for.
Exanple:
• Use Case: Customer Segmentation
• Explanation: An unsupervised learning algorithm can analyze
customer purchase data without predefined categories. It
may discover natural groupings or segments of customers
based on their buying behavior, helping businesses tailor
marketing strategies for different customer groups.
Reinforcement Learning:
Definition: Reinforcement learning involves training an agent to
make decisions in an environment by providing feedback in the
form of rewards or penalties. The agent learns to maximize
cumulative rewards over time by taking actions that lead to
positive outcomes.
Exanple:
• Use Case: Game Playing (e.g., Chess or Go)
• Explanation: A reinforcement learning algorithm can learn to
play a game by taking actions (moves) in an environment
(game board) and receiving rewards or penalties based on the
outcomes. The algorithm adapts its strategy over time to
maximize the chances of winning.
 Scenario 1:
A bank uses a model to detect fraudulent transactions. The model is
trained with a dataset of transactions that are labeled as either
"fraudulent" or "non-fraudulent."
Scenario 2:
A marketing department uses an algorithm to segment their customers
into different groups based on purchasing behavior, without any prior
labels on the groups.
Machine Scenario 3:
navigate
A roboticsvarious
companyterrains. The robot
develops receives
a model rewards
that helps for reaching
a robot learn to certain
points quickly and penalties for slow navigation.
Scenario 4:
A streaming service uses a model to predict what movies a user will like,
based on historical data of ratings given by the user to other movies.
Scenario 5:
An AI application analyzes social media data to discover underlying
patterns and group similar posts together, without any initial
categorization provided by humans.
 Answer to Scenario 1:
Type of Learning: Supervised Learning
Explanation: The model is trained with labeled data, distinguishing between fraudulent and
non-fraudulent transactions.
Answer to Scenario 2:
Type of Learning: Unsupervised Learning
Explanation: The algorithm identifies patterns and groups customers based on purchasing
behavior without pre-existing labels.

Answers and Answer to Scenario 3:

Type of Learning: Reinforcement Learning
Explanation: The robot learns to optimize its path and behavior through rewards and penalties,
adjusting its actions based on environmental feedback.
Answer to Scenario 4:
Type of Learning: Supervised Learning
Explanation: The model uses historical data of user ratings, which are explicit labels, to predict
user preferences.
Answer to Scenario 5:
Type of Learning: Unsupervised Learning
Explanation: The AI discovers patterns and groups in social media posts without any guided
input or labels, clustering similar content together.
Deep Learning

In deep learning, the algorithm learns to represent data in multiple levels of abstraction by using complex
architectures composed of interconnected layers. These layers transform the input data into increasingly
higher-level representations, ultimately leading to the desired output. Deep learning models are typically
trained on large datasets, often without explicit feature engineering, allowing them to automatically learn
relevant features from the data.

Exanple:
Use Case: Image Recognition
Explanation: In image recognition, a deep learning model can be trained on a dataset of images labeled
with corresponding objects or categories. The model learns to identify patterns and features within the
images by iteratively adjusting the parameters of its interconnected layers. Once trained, the deep
learning model can accurately classify new, unseen images, even when they contain variations in
lighting, perspective, or background clutter, making it a powerful tool for tasks like object detection and
image classification.
Natural Language
Processing
NLP stands for Natural Language Processing. It is a branch of artificial intelligence (AI) that
focuses on the interaction between computers and humans using natural language. The goal
of NLP is to enable computers to understand, interpret, and generate human language in a
way that is both meaningful and contextually relevant.

• Text Understanding: Extracting meaning from written or spoken language, understanding

syntax, semantics, and context.
• Speech Recognition: Converting spoken language into text.
• Text Generation: Creating coherent and contextually appropriate text based on a given
input.
• Machine Translation: Translating text or speech from one language to another.
• Naned Entity Recognition (NER): Identifying and classifying entities (such as names of
people, places, organizations) in text.
• Sentinent Analysis: Determining the sentiment or emotional tone expressed in a piece of
text.
• Text Sunnarization: Generating concise and informative summaries of longer pieces of
text.