Input: a set of solvents, X = {x1, ..., xn}

Output: a set of clusters, Y = {y1, ..., yn}

Begin

Initialize Y = {y1, ..., yn} randomly

Repeat select x ∈ X randomly

Find y* such that d(x,y*) = min{d(x,y)| y ∈ Y}

For all y ∈ N(y*) do

y = y + γ (x− y)

Reduce learning rate γ

Until termination condition is true

End