Unsupervised Learning - Kmeans Week 11 Challenge

Ans 1. When the data is categorical, traditional similarity measures like Euclidean distance are not suitable because they rely on numeric values.

Here are some common techniques:

a) Binary (0/1) Encoding:- Convert categorical variables into binary vectors using one-hot encoding or label encoding. Then, calculate similarity using:

b) Distance-based measures:- Calculate distance between categorical values using Hamming Distance.