Measuring Distance Between Data Points in Multidimensional Feature Space
5 types of distance functions (measures) used in machine learning algorithms
Distance plays an important role when discussing the similarity between data points. The lesser the distance between two data points, the more similar they are!
Many machine learning algorithms use distance functions to measure the similarity between data points. The dimension of the data doesn't matter here, distance functions work with any dimension, but we are only familiar with up to three dimensions. It is hard to imagine anything beyond that!
The algorithm's performance heavily depends on the type of distance function we use there. Here, we’ll discuss five types of distance functions used in machine learning algorithms. We begin with Euclidean distance, the most popular one!
1. Euclidean Distance
This is the most commonly used distance function which measures the shortest distance between two data points. In other words, it measures a straight line between two data points in a 2D plane.
