# Idea If two vectors/variables have been mean-centered (i.e., they are deviations from the mean), then the dot product is actually the [[covariance]]. Both measure "similarity" of two vectors. $ \operatorname{cov}_{x, y}=\frac{\sum{x_i y_i}}{N-1} $ $ \frac{\mathbf{x}^{\prime} \mathbf{y}}{N-1} $ # References