Abstract. This paper extends and generalizes the Bayesian semisupervised segmentation algorithm [1] for oil spill detection using SAR images. In the base algorithm on which we buil...
Traits are basically mixins or interfaces but with method bodies. In languages that support traits, classes are composed out of traits. There are two main advantages with traits. ...
This paper presents a novel approach of applying both positive selection and negative selection to supervised learning for anomaly detection. It first learns the patterns of the n...
A natural way to deal with training samples in imbalanced class problems is to prune them removing redundant patterns, easy to classify and probably over represented, and label noi...
Abstract. Building user-friendly GUIs for browsing and filtering RDF/S description bases while exploiting in a transparent way the expressiveness of declarative query/view language...