What is Machine Learning?

Machine Learning (ML) is the technology behind recommendation systems (Netflix, YouTube), spam filters (Gmail), voice assistants (Alexa, Siri), and self-driving cars. Let's understand what it is in simple words.

---

1. Traditional Programming vs Machine Learning

Traditional Programming (What you have been doing so far):

You write RULES (if-else logic) + INPUT DATA → Computer gives OUTPUT

Example: "If temperature > 37, print fever. Else print normal."

Machine Learning (A new way of thinking):

You give INPUT DATA + CORRECT ANSWERS → Computer learns the RULES automatically

Example: You give 10,000 patient records (temperature, symptoms, diagnosis). The computer finds the pattern and can predict diagnoses for new patients!

---

2. A Simple Analogy

Imagine you are teaching a child to recognize fruits:

• Traditional Programming: You write a rule book: "If it is round and red, it is an apple. If it is yellow and curved, it is a banana." (You write every rule manually.)

• Machine Learning: You show the child 100 pictures of apples and 100 pictures of bananas. After enough examples, the child learns the pattern and can identify new fruits on their own — even ones they have never seen before!

---

3. Types of Machine Learning

Supervised Learning (Most Common — This is what we will learn)

• You give the computer labeled data — data where you already know the correct answer.
• The computer learns the relationship between input and output.
• Example: Predicting house prices based on size, location, and number of rooms.

Unsupervised Learning

• You give the computer unlabeled data — data without answers.
• The computer finds hidden groups or patterns on its own.
• Example: Grouping customers into segments based on shopping behavior.

Reinforcement Learning

• The computer learns by trial and error, getting rewards for good actions.
• Example: Training a robot to walk or an AI to play chess.

For this course, we will focus on Supervised Learning because it is the most practical and commonly used in industry jobs.

---

4. Features (X) and Target (y)

In supervised learning, your data has two parts:

• Features (X): The information used to make predictions (also called "inputs" or "independent variables")
• Target (y): The answer you want to predict (also called "output" or "label")

Example: Predicting Student Pass/Fail

Study Hours (X₁)	Attendance % (X₂)	Past Score (X₃)	Result (y)
8	90	75	Pass
2	40	35	Fail
6	85	65	Pass
1	30	20	Fail

• Features (X): Study Hours, Attendance %, Past Score
• Target (y): Result (Pass/Fail)

The ML model learns the pattern: "Students who study more, attend regularly, and have good past scores tend to pass."

---

5. The 4 Steps of Every ML Project

No matter what ML model you build, the steps are always the same:

Step 1: PREPARE DATA    → Load, clean, split into X and y
Step 2: CHOOSE MODEL    → Pick an algorithm (Linear Regression, Decision Tree, etc.)
Step 3: TRAIN MODEL     → Feed data to the model using .fit(X, y)
Step 4: PREDICT & TEST  → Test with new data using .predict(X_new)

We will follow these exact 4 steps in the next lessons.

---

Summary

• Traditional programming = You write the rules. Machine Learning = The computer learns the rules from data.
• Supervised Learning = Learning from labeled data (input + correct answer).
• Features (X) = The input data. Target (y) = The answer to predict.
• Every ML project follows: Prepare → Choose → Train → Predict.