Kendall rank correlation coefficient

Ever wonder how to measure if two rankings are similar?

Imagine you're comparing the rankings of your favorite books from two different lists. You want to know how closely the rankings match.

Think of it as counting how many times two lists agree on the order of items. The closer the agreement, the stronger the connection between the lists.

Example

List A ranks "Book X" as 1st, "Book Y" as 2nd, and "Book Z" as 3rd. List B ranks "Book Y" as 1st, "Book X" as 2nd, and "Book Z" as 3rd. They agree on the order of "Book Y" and "Book Z", showing a strong connection.

Remember this

Spearman's rank correlation coefficient (ρ) quantifies the degree to which two rankings agree.

Text adapted from Wikipedia, licensed under CC BY-SA 4.0.

Related concepts

PageRank

PageRank formula: PR(A) = (1-d) + d Σ(PR(C)/L(C))

Pearson correlation coefficient

Pearson correlation coefficient formula: r = Σ[(xi - x̄)(yi - ȳ)] / [√(Σ(xi - x̄)²) * √(Σ(yi - ȳ)²)]

Phi coefficient

Matthews correlation coefficient (MCC) measures balanced metric even with class imbalance

score matching does: learns the gradient of the log-density without normalizing

Ever wonder how we can compare apples and oranges fairly in studies?

Evaluation of binary classifiers

Why can't we just use one score for everything?

Algorithms for calculating variance

Ever wondered how scientists measure the spread of data points?

Swipe through more Machine Learning concepts

Open Pocket Polymath