K-Nearest Neighbours Explained

K-Nearest Neighbours, often called KNN, classifies a new data point by looking at the categories of the closest existing data points.

If most of its nearest neighbors belong to one category, the new point gets assigned to that same category.

For example, if a new customer's spending pattern closely matches five existing customers, and four of them are labeled "high value," KNN would likely classify the new customer the same way.

• The value of K determines how many neighbours it checks before deciding.
• A small K can be sensitive to noise, while a large K can blur category boundaries.
• It works well on smaller datasets but slows down significantly on very large ones.

KNN is simple and intuitive because it mirrors how humans often judge things: by comparing them to similar known examples.