The note is for Dietterich, T. and Bakiri, G. (1995). Solving multiclass learning problems via error-correcting output codes, Journal of Artificial Intelligence Research 2: 263–286..
Approaches to multiclass learning problems
- direct application of multiclass algorithms, such as C4.5 and CART
- application of binary concept learning algorithms to learn individual binary functions for each of the $k$ classes
- application of binary concept learning algorithms with distributed output representations.
Error-correcting codes are employed as a distributed output representation
It is robust with respect to
- changes in the size of the training sample
- assignment of distributed representations to particular pruning 3, application of overfitting avoidance techniques such as decision-tree pruning.
And it can provide reliable class probability estimates.
- decision-tree methods, such as C4.5 and CART.
- artificial neural network algorithms, such as the perceptron algorithm and the error BP algorithm.
- direct multiclass approach: generalize decision-tree algorithms
- one-per-class approach: learn one binary function for each class with connectionist algorithms. ($f_i,i=1,2,\ldots,k$)
- distributed output code.