Makes use of algorithms that input unlabeled training data to attempt to deduce all the categories. Unsupervised learning is further divided into two subcategories, namely, clustering and dimensionality reduction. Which term describes this approach?

Makes use of algorithms that input unlabeled training data to attempt to deduce all the categories. Unsupervised learning is further divided into two subcategories, namely, clustering and dimensionality reduction. Which term describes this approach?

• A. Clustering and Dimensionality Reduction
• B. Classification
• C. Regression
• D. Supervised Learning

Answer: (A)

Discovering structure in data without labels is the hallmark of unsupervised learning. When you feed unlabeled data and try to infer natural groupings, you’re looking at clustering, which aims to assign data into categories based on similarity. Dimensionality reduction is the other main unsupervised technique that simplifies data while preserving its essential structure, helping reveal latent groups and relationships. The option that lists both clustering and dimensionality reduction best captures the described approach because it refers to the two primary unsupervised subfields that deal with discovering categories and structure from unlabeled data. The other terms describe supervised methods that require labeled examples to learn mappings or predict values.

Discovering structure in data without labels is the hallmark of unsupervised learning. When you feed unlabeled data and try to infer natural groupings, you’re looking at clustering, which aims to assign data into categories based on similarity. Dimensionality reduction is the other main unsupervised technique that simplifies data while preserving its essential structure, helping reveal latent groups and relationships. The option that lists both clustering and dimensionality reduction best captures the described approach because it refers to the two primary unsupervised subfields that deal with discovering categories and structure from unlabeled data. The other terms describe supervised methods that require labeled examples to learn mappings or predict values.