Summary: Proceedings of Engineering of Intelligent Systems '98 Conference (Ed. E Alpaydin), Vol 2, 6--12, ICSC Press, 1998.
Techniques for Combining Multiple Learners
Ethem Alpaydin
Department of Computer Engineering
Bo–gazi¸ci University
TR­80815 Istanbul, Turkey
alpaydin@boun.edu.tr
Abstract---Learners based on different paradigms can
be combined for improved accuracy. Each learning
method assumes a certain model that comes with a set
of assumptions which may lead to error if the assump­
tions do not hold. Learning is an ill­posed problem and
with finite data each algorithm converges to a different
solution and fails under different circumstances. Our pre­
vious experience with statistical and neural classifiers was
that classifiers based on these paradigms do generalize dif­
ferently, fail on different patterns and to a certain extent
complement each other and thus we look for ways to com­
bine them for higher accuracy. One way to get comple­
mentary classifiers is by using different input representa­

Source: Alpaydýn, Ethem - Department of Computer Engineering, Bogaziçi University

Collections: Computer Technologies and Information Sciences