Data Mining Techniques

Unit Structure

            
Data Mining Techniques
|
├── 1. Association Rules
│   ├── From Transaction Databases
│   ├── From Relational Databases
│   └── Correlation Analysis
│
├── 2. Classification and Prediction
│   └── Using Decision Tree Induction
│
└── 3. Clustering Techniques
    ├── Introduction to Clustering
    ├── Partition Method
    └── Hierarchical Method
            
        

• Once you understand what data mining is, the next step is learning how to actually discover patterns in data — and that’s where Data Mining Techniques come in. These techniques are like different tools in your toolbox, each made for digging out a specific kind of insight.
• We’ll start with Association Rules, which help uncover interesting relationships between items — like figuring out that people who buy bread often buy butter too. You’ll learn how these rules are mined from transactional databases (like shopping records) and relational databases, and how to measure the strength of these relationships using correlation analysis.
• Next, we move on to Classification and Prediction, where you’ll see how data can be used to predict outcomes — for example, classifying emails as spam or not spam. Here, we focus on a popular method called decision tree induction, which builds flowchart-like models for decision making.
• Finally, you’ll dive into Clustering Techniques, where the goal is to group similar data points together. You’ll explore the basics of clustering, and two main approaches: the partition method (like k-means clustering) and the hierarchical method (which builds nested clusters).