Source Of Summary:

Handout 5 - supervised learning.pdf

1. Supervised Learning 🎯

• K-Nearest Neighbor: Title of this handout. ✅
• Supervised Learning: The learning paradigm covered. 🔄

2. Classification vs Clustering 🤖

• Classification (known categories): Assigns each example to one of a fixed set of classes. 🎯
  • Also called Recognition or Supervised Classification.
• Clustering (unknown categories): Groups data into clusters without pre-specified labels. 🔍
  • Also called Unsupervised Classification.

3. Pattern Recognition Tasks 📚

• Classification:
  • Definition: Given a training set of records, each with attributes and one class attribute, learn a model for the class as a function of other attributes.
• Clustering:
  • Definition: Given data points with attributes and a similarity measure, find clusters so that points within a cluster are similar, and points across clusters are dissimilar.
• Other related tasks:
  • Association Rule Discovery: Given records each containing items, produce dependency rules predicting one item’s occurrence from others.
• An Early Task — Regression Analysis:
  • Definition: Statistical process for estimating relationships among variables; includes modeling and analyzing multiple variables.

4. Pattern Recognition Applications 🌐

• Wide range: Text recognition, image classification, bioinformatics, market segmentation, etc. ❗