Skip to content

Latest commit

 

History

History
85 lines (60 loc) · 3.35 KB

supervised_unsupervised_learning.md

File metadata and controls

85 lines (60 loc) · 3.35 KB

Supervised and Unsupervised Learning: Overview and Algorithms

1. Introduction to Machine Learning

Machine Learning is a subset of Artificial Intelligence that enables systems to learn from data and make predictions or decisions without being explicitly programmed.

Machine Learning is broadly categorized into:

  • Supervised Learning
  • Unsupervised Learning

2. Supervised Learning

In Supervised Learning, the model is trained on a labeled dataset, meaning that each training example is paired with an output label. The model learns to map inputs to the correct output.

Key Algorithms:

  1. Linear Regression

    • Definition: Predicts a continuous value based on the linear relationship between input variables.
    • Use Case: Predicting house prices based on size, location, etc.

  2. Logistic Regression

    • Definition: Used for binary classification problems.
    • Use Case: Email spam detection (spam or not spam).

  3. Decision Trees

    • Definition: Tree-like model of decisions, splits data into branches to reach an output.
    • Use Case: Customer churn prediction.

  4. Random Forest

    • Definition: Ensemble of decision trees to improve accuracy and avoid overfitting.
    • Use Case: Loan approval prediction.

  5. Support Vector Machines (SVM)

    • Definition: Finds the hyperplane that best separates classes.
    • Use Case: Face detection.

  6. K-Nearest Neighbors (KNN)

    • Definition: Classifies based on the majority label of nearest neighbors.
    • Use Case: Handwriting recognition.

  7. Naive Bayes

    • Definition: Probabilistic classifier based on Bayes' Theorem.
    • Use Case: Sentiment analysis.

3. Unsupervised Learning

In Unsupervised Learning, the data is not labeled. The algorithm tries to learn the underlying structure of the data.

Key Algorithms:

  1. K-Means Clustering

    • Definition: Partitions data into K distinct clusters.
    • Use Case: Customer segmentation in marketing.

  2. Hierarchical Clustering

    • Definition: Builds a hierarchy of clusters using a tree structure.
    • Use Case: Document classification.

  3. Principal Component Analysis (PCA)

    • Definition: Reduces the dimensionality of data while preserving most variance.
    • Use Case: Image compression.

  4. DBSCAN (Density-Based Spatial Clustering of Applications with Noise)

    • Definition: Finds core samples in dense regions and expands clusters from them.
    • Use Case: Anomaly detection in credit card transactions.

4. Key Differences Between Supervised and Unsupervised Learning

Feature Supervised Learning Unsupervised Learning
Data Labels Requires labeled data No labeled data required
Goal Predict outcomes Discover patterns or structure
Example Predicting stock prices Market basket analysis
Output Classification or Regression Clustering or Dimensionality Reduction

5. Conclusion

Understanding the difference between Supervised and Unsupervised Learning and their respective algorithms is critical for choosing the right approach in real-world applications. Supervised learning is suitable for prediction tasks, while unsupervised learning is ideal for discovering hidden patterns.