Some bounds of loss in classical semideterministic immunization

Since Samuelson, Redington and Fisher and Weil, duration and immunization are very important topics in bond portfolio analysis from both a theoretical and a practical point of view. Many results have been established, especially in semi-deterministic framework. As regards, however, the loss may be sustained, we do not think that the subject has been investigated enough, except for the results found in the wake of the theorem of Fong and Vasicek. In this paper we present some results relating to the limitation of the loss in the case of local immunization for multiple liabilities.

O livro na passagem de “átomos” a “bits”: as tecnologias digitais no contexto do livro electrónico

As Tecnologias de Informação e Comunicação colocam em debate o modo de pensar e olhar o livro tradicional e originam novas materialidades para o texto e novas formas, espaços e géneros de leitura. A revolução originada por Gutenberg democratizou o acesso ao livro tradicional, alterando os processos de acesso ao conhecimento. As tecnologias eletrónicas trouxeram consigo uma nova revolução, virtualizando o acesso aos textos que alteraram a paisagem da cultura clássica do livro. Recorrendo à ideia original do investigador Nicholas Negroponte, pretende-se realizar uma análise das alterações que o livro sofreu, enquanto objeto de suporte ao texto, na sua passagem para contextos digitais ou na passagem de “átomos” para “bits”.

Esto es amor: trabalho de projeto

Mestrado Teatro, especialização em artes performativas, teatro música

Text classification using compression-based dissimilarity measures

Arguably, the most difficult task in text classification is to choose an appropriate set of features that allows machine learning algorithms to provide accurate classification. Most state-of-the-art techniques for this task involve careful feature engineering and a pre-processing stage, which may be too expensive in the emerging context of massive collections of electronic texts. In this paper, we propose efficient methods for text classification based on information-theoretic dissimilarity measures, which are used to define dissimilarity-based representations. These methods dispense with any feature design or engineering, by mapping texts into a feature space using universal dissimilarity measures; in this space, classical classifiers (e.g. nearest neighbor or support vector machines) can then be used. The reported experimental evaluation of the proposed methods, on sentiment polarity analysis and authorship attribution problems, reveals that it approximates, sometimes even outperforms previous state-of-the-art techniques, despite being much simpler, in the sense that they do not require any text pre-processing or feature engineering.