Abstract

Abstract

Abstract
Every year thousands of students apply for access to higher education in Portuguese universities. These students are allocated according to their application, secondary school grades and their preference to distinct establishment-course pairs. Despite official statistics provided by the Ministry of Science, Technology and Higher Education (MCTES), it is not always possible to have the relevant indicators for universities make a good decision concerning courses and vacancies to offer. This study evaluates the factors that influence students in their choice of a pair establishment-course as well as understand the different variables that are related to the same attribution of influence. The study is observational and empirical using questionnaires, with the appropriate statistical analysis. The results showed a relation in greater number among the studied factors and gender. In conclusion the prestige of the institution, employment and intellectual development are the factors most valued by students. © 2014 AISTI.

Abstract
Dynamic real-world applications that generate data continuously have introduced new challenges for the machine learning community, since the concepts to be learned are likely to change over time. In such scenarios, an appropriate model at a time point may rapidly become obsolete, requiring updating or replacement. As there are several learning algorithms available, choosing one whose bias suits the current data best is not a trivial task. In this paper, we present a meta-learning based method for periodic algorithm selection in time-changing environments, named MetaStream. It works by mapping the characteristics extracted from the past and incoming data to the performance of regression models in order to choose between single learning algorithms or their combination. Experimental results for two real regression problems showed that MetaStream is able to improve the general performance of the learning system compared to a baseline method and an ensemble-based approach.

Abstract
Let f and g be smooth multimodal maps with no periodic attractors and no neutral points. If a topological conjugacy h between f and g is C-1 at a point in the nearby expanding set of f, then h is a smooth diffeomorphism in the basin of attraction of a renormalization interval of f. In particular, if f:I -> I and g : J -> J are C-r unimodal maps and h is C-1 at a boundary of I, then h is C-r in I.

Abstract