Unit 2: Techniques for Supervised and Unsupervised Learning

Question 1

What is supervised learning, and what are its key characteristics?

Answer 1

Supervised Learning: A type of machine learning where the model is trained on labeled data (input-output pairs).
Key Characteristics:
Labeled Data: Each training example is paired with an output label.
Objective: To learn a mapping from inputs to outputs.
Applications: Classification (e.g., spam detection) and regression (e.g., predicting prices).

Question 2

List common algorithms used in supervised learning and their applications.

Answer 2

Common Algorithms:

Linear Regression: Used for predicting continuous outcomes.
    Example: Predicting housing prices based on features like size and location.
Logistic Regression: Used for binary classification problems.
    Example: Predicting whether an email is spam or not.
    Equation: P(Y=1∣X)=11+e−(β0+β1X)P(Y=1∣X)=1+e−(β0​+β1​X)1​
Decision Trees: Used for both classification and regression tasks.
    Example: Classifying types of fruits based on attributes like color and weight.
Support Vector Machines (SVM): Effective for high-dimensional spaces in classification.
    Example: Image classification tasks.

Question 3

List common algorithms used in unsupervised learning and their applications.

Answer 3

Common Algorithms:

K-Means Clustering: Groups data points into k clusters based on feature similarity.
    Example: Segmenting customers into different groups based on purchasing behavior.
    Equation: Minimize ∑i=1k∑j=1n∣∣xj−μi∣∣2∑i=1k​∑j=1n​∣∣xj​−μi​∣∣2 (where μiμi​ is the centroid of cluster ii).
Hierarchical Clustering: Builds a hierarchy of clusters using either agglomerative or divisive methods.
    Example: Creating a dendrogram to visualize customer groupings.
Principal Component Analysis (PCA): Reduces dimensionality while preserving variance in data.
    Example: Reducing the number of features in a dataset to visualize data in two dimensions.

Question 4

What is unsupervised learning, and how does it differ from supervised learning?
A:

Answer 4

Unsupervised Learning: A type of machine learning where the model is trained on unlabeled data.
Key Differences from Supervised Learning:

Lack of Labeled Data: No explicit output labels for training examples.
Objective: To identify patterns or groupings within the data.
Applications: Clustering (e.g., customer segmentation) and association (e.g., market basket analysis).