Abstract— Automatic pattern classifiers that allow for incremental learning can adapt internal class models efficiently in response to new information, without having to retrai...
This research introduces a general class of functions serving as generalized neuron models to be used in artificial neural networks. They are cast in the common framework of comp...
Abstract. This paper outlines a strategy to build new bilingual dictionaries from existing resources. The method is based on two main tasks: first, a new set of bilingual correspo...
Implementing the 23 Gang-of-Four design patterns in the aspectoriented programming language Object Teams/Java (OT/J) yields modularity and reusability results roughly comparable t...
We study the potential benefits to classification prediction that arise from having access to unlabeled samples. We compare learning in the semi-supervised model to the standard, ...