AI Path Finder
Free consultation

Lesson · Hinglish · ⏱ 6 min padhne mein

🎯 Do tarah ki prediction: regression aur classification

EnglishHinglish

🏠 Asli zindagi ki misaal

Swiggy app do alag tarah ke sawaalon ke jawaab deta hai. 'Delivery mein kitne minute lagenge?' ka jawaab ek number hai: yeh regression hai. 'Kya yeh order late hoga: haan ya na?' order ko category mein daalta hai: yeh classification hai.

💡 Aasaan shabdon mein

Regression ek continuous number predict karta hai, jaise price, temperature ya delivery time. Linear regression, jo classic starting point hai, features aur number ke beech sabse acchi seedhi line ka rishta banata hai. Classification ek category predict karta hai, jaise spam/not spam, fraud/genuine, ya 10 product types mein se kaunsa. Logistic regression, naam ke bawajood, ek simple classification model hai jo probability deta hai. Decision trees dono kar sakte hain. Kuch complex try karne se pehle hamesha in simple models se shuru karo.

🪜 Step by step

  1. Poochho jawaab kaisa dikhta hai. Number (keemat, delivery din) matlab regression. Category (spam ya nahi, kaunsa product type) matlab classification.
  2. Regression ke liye linear regression se shuru karo; error asli units mein naapo (MAE: average ₹1,200 ka fark).
  3. Classification ke liye logistic regression se shuru karo; yeh probability deta hai jis pe threshold laga sakte ho.
  4. Ek seedhe baseline se compare karo: hamesha average predict karo, ya hamesha sabse common class.
  5. Phir strong models try karo (random forest, gradient boosting) aur sabse simple jo saaf jeete use rakho.

🔑 Zaroori shabd

Regression
Number predict karna, jaise ghar ka price.
Classification
Category predict karna, jaise spam hai ya nahi.
Baseline
Sabse simple prediction, jise aapke model ko beat karna hai.

👀 Example dekho

python
from sklearn.linear_model import LinearRegression, LogisticRegression

reg = LinearRegression().fit(X_train, delivery_minutes)   # regression
clf = LogisticRegression().fit(X_train, is_late)           # classification

reg.predict(X_new)          # → [32.5] minutes
clf.predict_proba(X_new)    # → [[0.8, 0.2]] → 20% chance of being late

Same features, do alag sawaal, do alag model types.

⚠️ Aam galtiyaan jinse bachna hai

  • Categories ko numbers maan lena, jaise city codes 1, 2, 3, to model sochta hai city 3 city 1 se 'zyaada' hai. One hot encoding use karo.
  • Regression ke liye accuracy ya classification ke liye MAE use karna. Har type ke apne metrics hain.

✅ Jaldi se check karo

Kal website pe kitne visitors aayenge, yeh predict karna kya hai?

  1. Regression
  2. Clustering
  3. Classification
Jawaab dekho

A. Regression: Jawaab ek number hai (visitors ki ginti), isliye yeh regression hai.

🙋 Seekhne wale yeh poochhte hain

Classify karta hai to logistic 'regression' kyun kehte hain?

Purana naam hai: yeh ek number, probability, predict karta hai, jise threshold se class mein badla jaata hai.

Kya ek model dono kar sakta hai?

Kai algorithms ke dono versions hote hain, jaise decision trees aur gradient boosting. Target ke hisaab se version chuno.

📌 Yaad rakho

Number → regression. Category → classification. Simple se shuru karo.

📖 Is lesson ke shabd (AI dictionary)

Poora course, aapke roadmap ke hisaab se

Progress tracking, quick checks aur step-by-step order ke saath. Yeh lesson in courses ka hissa hai:

AI Foundations Path · ML Engineer

Course mein seekho