MultiBoosting: A technique for combining boosting and wagging

MultiBoosting is an extension to the highly successful AdaBoost technique f or forming decision committees. MultiBoosting can be viewed as combining Ad aBoost with wagging. It is able to harness both AdaBoost's high bias and va riance reduction with wagging's superior variance reduction. Using C4.5 as the base learning algorithm, MultiBoosting is demonstrated to produce decis ion committees with lower error than either AdaBoost or wagging significant ly more often than the reverse over a large representative cross-section of UCI data sets. It offers the further advantage over AdaBoost of suiting pa rallel execution.