linearly separable

Page numbers are for draft copy at present; they will be replaced with correct numbers when final book is formatted. Chapter numbers are correct and will not change now.

Two sets of data are linearly separable if there is a straight line (or hyperplane in higher dimensions) that separates them. In other words there is a linear combination (a_i) of the features and a threshold T such that fp
      a₁d₁ + a₂d₂ + ... + a_nd_n is less than T for data items (d_i) in the first datasets and greater than T for the second
In such cases, the hyperplane acts as a {[classifier}}.

Used on Chap. 6: pages 111, 112; Chap. 7: page 136

Also known as linear separability