DMCT-NOTES/unit 1/03_Data_Mining_Techniques.md

Data Mining Techniques

There are several key techniques used to mine data.

1. Classification (Predictive)

• Goal: Assign items to predefined categories (classes).
• Supervised Learning: We know the categories beforehand.
• Example: Is this email Spam or Not Spam?

2. Regression (Predictive)

• Goal: Predict a continuous number.
• Example: Predicting the price of a house based on its size and location.

3. Clustering (Descriptive)

• Goal: Group similar items together.
• Unsupervised Learning: We don't know the groups beforehand.
• Example: Grouping customers into segments (e.g., "High Spenders", "Budget Shoppers").

4. Association Rules (Descriptive)

• Goal: Find relationships between items.
• Market Basket Analysis: "People who buy Bread often also buy Butter."
• Key Terms:
  • Support: How often items appear together.
  • Confidence: How likely item B is purchased if item A is purchased.

5. Outlier Detection

• Goal: Find unusual data points that don't fit the pattern.
• Example: Detecting credit card fraud (a huge transaction in a usually quiet account).