Percorrer por autor "Sousa, Ricardo"
A mostrar 1 - 2 de 2
Resultados por página
Opções de ordenação
- Multicriteria Models for Learning Ordinal Data: A Literature ReviewPublication . Sousa, Ricardo; Yevseyeva, Iryna; Pinto da Costa, Joaquim F.; Cardoso, Jaime S.Operations Research (OR) and Artificial Intelligence (AI) disciplines have been playing major roles on the design of new intelligent systems. Recently, different contributions from both fields have been made on the models design for problems with multi-criteria. The credit scoring problem is an example of that. In this problem, one evaluates how unlikely a client will default with his payments. Client profiles are evaluated, being their results expressed in terms of an ordinal score scale (Excelent Good Fair Poor). Intelligent systems have then to take in consideration different criteria such as payment history, mortgages, wages among others in order to commit their outcome. To achieve this goal, researchers have been delving models capable to render these multiple criteria encompassed on ordinal data. The literature presents a myriad of different methods either on OR or AI fields for the multi-criteria models. However, a description of ordinal data methods on these two major disciplines and their relations has not been thoroughly conducted yet. It is key for further research to identify the developments made and the present state of the existing methods. It is also important to ascertain current achievements and what the requirements are to attain intelligent systems capable to capture relationships from data. In this chapter one will describe techniques presented for over more than five decades on OR and AI disciplines applied to multi-criteria ordinal problems.
- Testing the Maximum by the Mean in Quantitative Group TestsPublication . Martins, João Paulo; Santos, Rui; Sousa, RicardoGroup testing, introduced by Dorfman in 1943, increases the efficiency of screening individuals for low prevalence diseases. A wider use of this kind of methodology is restricted by the loss of sensitivity inherent to the mixture of samples. Moreover, as this methodology attains greater cost reduction in the cases of lower prevalence (and, consequently, a higher optimal batch size), the phenomenon of rarefaction is crucial to understand that sensitivity reduction. Suppose, with no loss of generality, that an experimental individual test consists in determining if the amount of substance overpasses some prefixed threshold l. For a pooled sample of size n, the amount of substance of interest is represented by (Y1, … , Yn), with mean (Formula Presented) and maximum Mn. The goal is to know if any of the individual samples exceeds the threshold l, that is, Mn > l. It is shown that the dependence between (Formula Presented) and Mn has a crucial role in deciding the use of group testing since a higher dependence corresponds to more information about Mn given by the observed value of (Formula Presented).