Sistemas Informáticos

There is still lack of clarity about the best manner in which to handle numeric attributes when applying Bayesian network classifiers. Discretization methods entail an unavoidable loss of information. Nonetheless, a number of studies have shown that appropriate discretization can outperform straightforward use of common, but often unrealistic parametric distribution (eg Gaussian).

Abstract AODE (Aggregating One-Dependence Estimators) is considered one of the most interesting representatives of the Bayesian classifiers, taking into account not only the low error rate it provides but also its efficiency. Until now, all the attributes in a dataset have had to be nominal to build an AODE classifier or they have had to be previously discretized. In this paper, we propose two different approaches in order to deal directly with numeric attributes.

Resumen—Son varios los motivos por los que es deseable llevar a cabo una reducción de la dimensionalidad y del número de instancias de la base de datos de interés en tareas de clasificación. Especial interés despierta el clasificador basado en los k vecinos más cercanos, por su gran demanda de recursos computacionales, tanto a nivel espacial como temporal en la clasificación.

RÉSUMÉ. L'AODE est consideré comme l'un des algorithmes representatifs les plus intéressants parmi les classifieurs bayésienes, si l'on considère non seulement son taux d'erreur peu élevé mais aussi son efficacité. Jusqu'à présent, pour pouvoir construire un classifieur de type AODE, tous les attributs (ou variables) dans un ensemble de données devaient être nominaux ou bien devaient être des variables préalablement discrétisées.

Abstract Task modeling has become one of the cornerstones of model-based user inter-face design. In this paper, a task-based approach to user interfaces design is in-troduced. This approach is supported by a tool, namely IDEALXML, that al-lows for the animation of the specified ...