This paper presents Fuzzy-UCS, a Michigan-style Learning Fuzzy-Classifier System designed for supervised learning tasks. FuzzyUCS combines the generalization capabilities of UCS w...
Albert Orriols-Puig, Jorge Casillas, Ester Bernad&...
Background: The receiver operating characteristic (ROC) curve is a fundamental tool to assess the discriminant performance for not only a single marker but also a score function c...
The Kushilevitz-Mansour (KM) algorithm is an algorithm that finds all the “large” Fourier coefficients of a Boolean function. It is the main tool for learning decision trees ...
: Boosting is a general method for improving the accuracy of any given learning algorithm. In this paper we employ combination of Adaboost with Support Vector Machine (SVM) as comp...
This paper introduces a Bayesian algorithm for constructing predictive models from data that are optimized to predict a target variable well for a particular instance. This algori...