Jh. Holmes et al., The learning classifier system: an evolutionary computation approach to knowledge discovery in epidemiologic surveillance, ARTIF INT M, 19(1), 2000, pp. 53-74
The learning classifier system (LCS) integrates a rule-based system with re
inforcement learning and genetic algorithm-based rule discovery. This inves
tigation reports on the design, implementation, and evaluation of EpiCS, a
LCS adapted for knowledge discovery in epidemiologic surveillance. Using da
ta from a large, national child automobile passenger protection program, Ep
iCS was compared with C4.5 and logistic regression to evaluate its ability
to induce rules from data that could be used to classify cases and to deriv
e estimates of outcome risk, respectively. The rules induced by EpiCS were
less parsimonious than those induced by C4.5, but were potentially more use
ful to investigators in hypothesis generation. Classification performance o
f C4.5 was superior to that of EpiCS (P < 0.05). However, risk estimates de
rived by EpiCS were significantly more accurate than those derived by logis
tic regression (P < 0.05). (C) 2000 Elsevier Science B.V. All rights reserv
ed.