Fisher 线性判别

　　Multiplying both sides of this result by w^T and adding w₀, and making use of y(x)=w^Tx+w₀ and  y(x_Γ)=w^Tx_Γ+w₀=0, we have r=y(x)/||w||.

　　The idea proposed by Fisher is to maximize a function that will give a large separation between the projected class means while also giving a small variance within each class, thereby minimizing the class overlap.