Scribd
Upload
2 views3 pages

Supervised 2

Supervised Learning is a type of machine learning where a model is trained using input data along with the correct output (label) to learn the relationship between them for making predictions. It involves a training phase where the model learns from labeled data, and crucial concepts include train-test split, overfitting, and underfitting. Common algorithms for supervised learning include classification methods like Logistic Regression and regression methods like Linear Regression.

Uploaded by

Rishabh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
2 views3 pages

Supervised 2

Supervised Learning is a type of machine learning where a model is trained using input data along with the correct output (label) to learn the relationship between them for making predictions. It involves a training phase where the model learns from labeled data, and crucial concepts include train-test split, overfitting, and underfitting. Common algorithms for supervised learning include classification methods like Logistic Regression and regression methods like Linear Regression.

Uploaded by

Rishabh
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

AIML Notes Supervised Learning

Supervised Learning
Machine Learning – Detailed Notes

1. What is Supervised Learning?


Supervised Learning Machine Learning ka wo type hai jisme model ko input data ke
saath correct output (label) diya jaata hai.
Model input aur output ke beech ka relation seekhta hai aur phir naye data par
prediction karta hai.
Input + Correct Output → Training → Model

2. Why is it called Supervised?


Is learning process me model ko pehle se bataya jaata hai ki:
• Ye input hai
• Ye iska correct output hai
Isliye learning guided hoti hai, bina guesswork ke.

3. Basic Training Example


Marks Result
30 Fail
50 Pass
70 Pass
90 Pass
Yahan:
• Marks = Input feature
• Result = Label (Output)
Model seekhta hai ki kis range ke marks par kaunsa output aata hai.

4. How Supervised Learning Works


1. Input data collect kiya jaata hai
2. Har input ke saath correct output diya jaata hai
3. Model training ke dauraan errors calculate karta hai
4. Errors gradually kam karta jaata hai
5. Trained model prediction ke liye ready hota hai
Input Data + Labels → Training → Model → Prediction

1
AIML Notes Supervised Learning

5. Crucial Concepts (Must Know)


Machine Learning start karne se pehle ye concepts samajhna zaroori hai.

5.1 1. Train-Test Split


Hum poora data training ke liye use nahi karte. Data ko do parts me divide kiya jaata
hai:

• Training Data (80%): Jisse model seekhta hai

• Testing Data (20%): Jise model ne pehle nahi dekha

Agar model ko same data par test kiya jaye jis par train hua tha, toh wo data ko
memorize kar lega. Isliye testing data real performance check karne ke liye use hota hai.

5.2 2. Overfitting vs Underfitting


Overfitting: Jab model training data ko bahut detail me yaad kar leta hai, par naye
data par perform nahi karta.

Model ne data rat liya, concept nahi samjha.

Underfitting: Jab model ne pattern theek se seekha hi nahi. Training aur testing
dono me poor performance hoti hai.
Good Fit: Jab model ne actual concept seekh liya ho aur naye data par bhi sahi
prediction kare.
(Yahan overfitting vs underfitting ka graph show kiya jaata hai.)

6. Types of Supervised Learning


6.1 1. Classification
Output categorical hota hai.

• Pass / Fail

• Spam / Not Spam

• Yes / No

6.2 2. Regression
Output numeric hota hai.

• Salary prediction

• House price prediction

• Temperature prediction

2
AIML Notes Supervised Learning

7. Common Supervised Learning Algorithms


7.1 Classification
Logistic Regression, KNN, Decision Tree, SVM, Naive Bayes

7.2 Regression
Linear Regression, Polynomial Regression, Decision Tree Regression

8. Notation (X and y)
Coding start karte waqt ye notation standard hota hai:

• X: Input features (independent variables)

• y: Output label (target variable)

model . fit ( X_train , y_train )


model . predict ( X_test )

9. One-Line Summary
Supervised Learning me model ko input ke saath correct output diya jaata hai, jisse wo
data se seekhkar accurate predictions karta hai.

You might also like