Unsupervised Learning

 

📘 Unsupervised Learning

🔹 Definition

Unsupervised Learning is a type of Machine Learning where the model is trained on unlabeled data, i.e., data without predefined outputs.

👉 Goal:
Discover hidden patterns, structures, or relationships in data.


This type of learning is particularly useful for discovering clusters in data, detecting anomalies, and finding associations. Techniques like clustering, association rule mining, and dimensionality reduction allow for deeper insights into complex datasets. It's widely used in customer segmentation, fraud detection, and market basket analysis.

🔹 Key Idea

Given only input data:

X={x1,x2,x3,...,xn}X = \{x_1, x_2, x_3, ..., x_n\}

👉 The model tries to:

  • Group similar data
  • Identify patterns
  • Reduce dimensionality

🔹 Characteristics

  • No labeled outputs
  • No “teacher”
  • Exploratory in nature
  • Used for pattern discovery


Pic Courtesy: database town

🔹Types of Unsupervised Learning

🟦 A. Clustering

📌 Definition

Clustering is the process of grouping data points into clusters based on similarity. It helps uncover hidden patterns and groupings within the data, which can be useful for customer segmentation, anomaly detection, or identifying trends.

Grouping data points into clusters such that:

  • Points in same cluster → similar
  • Points in different clusters → dissimilar

📊 Popular Algorithms

1. K-Means Clustering

  • Divides data into K clusters
  • Iteratively updates cluster centers

2. Hierarchical Clustering

  • Builds a tree (dendrogram)
  • Two types:
    • Agglomerative (bottom-up)
    • Divisive (top-down)

3. DBSCAN

  • Density-based clustering
  • Detects arbitrary shaped clusters


🟩 B. Association Rule Learning

📌 Definition

Finds relationships between variables in large datasets

📊 Key Algorithms

1. Apriori Algorithm

  • Finds frequent itemsets
  • Generates rules

2. FP-Growth

  • Faster alternative to Apriori

📌 Example:

Market Basket Analysis:

  • “Customers who buy bread also buy butter”

🟨 C. Dimensionality Reduction

📌 Definition

Reduce number of features while preserving important information

📊 Techniques

1. PCA (Principal Component Analysis)

  • Projects data to lower dimensions
  • Maximizes variance

2. t-SNE

  • Used for visualization

📌 Example:

Reducing 100 features → 2 features for visualization

🔹How It Works

  1. Input raw data (no labels)
  2. Measure similarity/distance
  3. Apply algorithm
  4. Discover structure

🔹Distance Measures (Important)

Used in clustering:

  • Euclidean Distance
  • Manhattan Distance
  • Cosine Similarity

🔹Example 1: Customer Segmentation

Input:

  • Age, income, spending

Output:

  • Customer groups

👉 Helps businesses target customers

📌 Detailed Explanation: Customer Segmentation

Customer segmentation is a common example of unsupervised learning.

Problem:

A company wants to group customers based on similarities in their purchasing behavior, without any predefined labels or categories.

Data Collected:

  • Age
  • Purchase history
  • Spending habits

Techniques Used:

The company applies unsupervised learning algorithms such as:

  • K-means clustering → to group similar customers
  • Principal Component Analysis (PCA) → to reduce data complexity

Key Methods & Insights

1. Clustering

  • Groups customers with similar behaviors into segments
  • Example segments:
    • High-spending customers
    • Budget-conscious buyers

2. Association Rule Mining

  • Finds relationships between purchases
  • Example:
    ➡️ Customers who buy product X are likely to buy product Y

3. Dimensionality Reduction

  • Simplifies the dataset by focusing on key features
  • Retains important patterns while reducing complexity
  • Example: focusing on age and spending habits

4. Anomaly Detection

  • Identifies unusual or rare behaviors
  • Example:
    ➡️ Customers with irregular or unexpected purchasing patterns

Key Idea:

➡️ Unsupervised learning discovers hidden patterns and structures in data without labeled outputs.

🔹 Example 2: Market Basket Analysis

Input:

  • Transaction data

Output:

  • Association rules

👉 Used in retail

🔹 Example 3: Anomaly Detection

  • Identify unusual patterns

👉 Example:

  • Fraud detection
  • Network intrusion detection

🔹Applications of Unsupervised Learning

🛒 Retail

  • Customer segmentation
  • Product recommendation

💰 Finance

  • Fraud detection
  • Risk analysis

🧬 Bioinformatics

  • Gene clustering
  • Protein analysis

📱 Social Media

  • Community detection
  • Trend analysis

🎥 Computer Vision

  • Image compression
  • Feature extraction

🔹 Advantages

  • No need for labeled data
  • Useful for exploratory analysis
  • Can discover unknown patterns

🔹 Disadvantages

  • Results may be hard to interpret
  • No clear evaluation metric
  • Sensitive to noise and parameters

🔹Comparison with Supervised Learning

Feature    Supervised Learning    Unsupervised Learning
Data    Labeled    Unlabeled
Goal    Prediction    Pattern discovery
Output    Known labels    Hidden structure

📝 Summary 

  • Unsupervised learning uses unlabeled data
  • Main tasks:
    • Clustering
    • Association rules
    • Dimensionality reduction
  • Used for:
    • Pattern discovery
    • Data exploration

Comments

Post a Comment

Popular posts from this blog

Machine Learning PCCST503 Semester5 KTU CS 2024 Scheme - Dr Binu V P

About Me  - Dr Binu V P Syllabus Model Question Paper Machine Learning PCCST503 Semester 5 KTU CS 2024 Scheme Do the programs to better understand the concept. Machine Learning Lab PCCSL508 Module -I Introduction to Machine Learning Introduction to ML Distinguishing Machine Learning from Traditional Programming Machine Learning Paradigms Supervised UnSupervised Semi-Supervised Reinforcement Stages of Machine Learning: From Data Collection to Deployment Parameter Estimation Introduction to Parameter Estimation in Machine Learning Maximum Likelyhood Estimation.(MLE) Maximum A Posteriori Estimation.(MAP) Numerical Problems MLE and MAP Supervised Learning Feature Representation and Problem Formulation Role of loss functions and optimization  Regression Introduction to Regression   Linear regression with one variable  (  lab experiment ) Example Problems- simple linear regression - University Questions Evaluation Metrics for Linear Regression Linear Regression...
Read more

Introduction to Machine Learning (ML)

  📘 Introduction to Machine Learning (ML) 🔹 What is Machine Learning? Machine Learning is a subfield of Artificial Intelligence that focuses on developing algorithms that allow computers to learn patterns from data and make decisions or predictions without explicit programming . It focuses on developing algorithms and models that allow machines to: Identify patterns in data Make decisions with minimal human intervention Predict future outcomes based on past experiences By continuously learning from new data, machine learning systems become more accurate and efficient over time. 📌 Formal Definition  A computer program is said to learn from experience E with respect to some task T and performance measure P if its performance at task T , as measured by P , improves with experience E . 🔹 Why Machine Learning? Traditional programming: Input + Program → Output Machine Learning paradigm: Input + Output → Model (learned program) 👉 Instead of writing...
Read more