Beatriz Lucía

La influencia del tutor en el seguimiento de programas eLearning: Estudio de acciones en un caso práctico

Resumen es: Las propuestas de Educacion a Distancia basadas en tecnologia Web ocupan en la actualidad un lugar indiscutible. El uso de los nuevos recursos en los pro...

Small sample discrimination and professional performance assessment

Class overlapping and small sample class sizes, a situation not infrequent in practical settings, can make very difficult the successful construction of a classification system. In this paper we will address this question by means of a new procedure, which we call Nonlinear Discriminant Analysis (NLDA), for classifier construction in such cases and that combines the excellent approximation properties of the well known Multilayer Perceptrons with the target-free classical discrimination technique of Fishers Analysis. Besides a short description of NLDA fundamentals, we will give an illustration of its use in a practical problem, the assessment of professional performance of insurance salespersons.