Unsupervised Learning - Kmeans Week 11 Challenge

1) Similarity in categorical data is not as direct when itcomes to numerical data. the categorical data is first tranformed to numerical values and both are standardised to make sense of the similarity and dissimilarity between/among those features.

Methods which is used to calculate the distance or similarities:

a)Hamming Distance: Imagine you have two lists of categories (red, blue, green). Hamming distance is just counting how many categories are different in the same position on the two lists.

b) Matching Coefficient:The matching coefficient is the ratio of the number of agreements (matching categories) to the total number of categories in both sets.

c)Categorical Distance Measures: There are various categorical distance measures, such as the Gower distance, which takes into account the mixed nature of data (both numerical and categorical).

d) Jaccard Similarity: In the context of categorical data, each category represents a set, and the Jaccard similarity is calculated as the size of the intersection divided by the size of the union of the sets.