KEMBAR78
Unsupervisedlearning | PDF
Scribd
Upload
17 views11 pages

Unsupervisedlearning

Unsupervised Learning is a machine learning approach that analyzes unlabeled data to discover patterns and structures. It primarily includes clustering and dimensionality reduction techniques, with popular algorithms like K-Means and PCA. Applications span various fields such as marketing, security, healthcare, and natural language processing.

Uploaded by

important9505
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
17 views11 pages

Unsupervisedlearning

Unsupervised Learning is a machine learning approach that analyzes unlabeled data to discover patterns and structures. It primarily includes clustering and dimensionality reduction techniques, with popular algorithms like K-Means and PCA. Applications span various fields such as marketing, security, healthcare, and natural language processing.

Uploaded by

important9505
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 11

Unsupervised Learning

Discovering Patterns Without Labels


Introduction

Unsupervised Learning is a type of machine learning where the


algorithm is trained on unlabeled data.

Unlike supervised learning, there are no input-output pairs.

The goal is to discover patterns, groupings, or hidden


structures in the data.

Key Idea:
“Let the machine figure out the structure of data by itself.”
• Real-Life Examples
• Market Segmentation: Grouping customers
based on buying behavior.
• Netflix & Spotify: Recommending shows/music
by clustering users with similar tastes.
• Anomaly Detection: Detecting fraud or unusual
activity in bank transactions.
• Image Segmentation: Grouping similar pixels in
images.
Types of Unsupervised Learning
• Types of Unsupervised Learning
• Clustering
– Grouping data points into clusters based on similarity.
– Algorithms:
• K-Means
• Hierarchical Clustering
• DBSCAN
• Dimensionality Reduction
– Reducing features while preserving important information.
– Algorithms:
• PCA (Principal Component Analysis)
• t-SNE
• Autoencoders
Advantages
• Works with unlabeled data
• Helps find hidden patterns
• Useful for visualization
Popular algorithms
• K-Means Clustering
• Groups data into K clusters.
• Example: Grouping students by their study habits.
• Steps:
– Choose K clusters.
– Assign points to nearest cluster center.
– Recalculate cluster centers.
– Repeat until stable.
• PCA (Principal Component Analysis)
• Reduces dimensionality of data.
• Example: A student dataset with 50 features can be reduced to
2–3 features for visualization.
Limitations
• Hard to evaluate
• Sensitive to noise
• May produce meaningless clusters
Applications
- Marketing: Customer segmentation
- Security: Anomaly detection
- Healthcare: Patient grouping
- NLP: Topic modeling
• Summary
• Unsupervised learning = finding patterns
without labels.
• Two main types: Clustering & Dimensionality
Reduction.
• Popular algorithms: K-Means, Hierarchical
Clustering, PCA, DBSCAN.
• Useful in exploration, segmentation, anomaly
detection, NLP, and vision.
Example: K-Means Clustering
Example: PCA (Dimensionality Reduction)

You might also like