Rio de Janeiro, Brazil
Rio de Janeiro, Brazil

Time filter

Source Type

Revoredo K.,Unirio | Ochoa-Luna J.E.,University of Sao Paulo | Cozman F.G.,University of Sao Paulo
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2010

This paper investigates learning methods where the target language is the recently proposed probabilistic description logic cr . We start with an inductive logic programming algorithm that learns logical constructs; we then develop an algorithm that learns probabilistic constructs by searching for conditioning concepts, using examples given as interpretations. Issues on learning from entailments are also examined, and practical examples are discussed. © 2010 Springer-Verlag.


Ochoa-Luna J.E.,University of Sao Paulo | Revoredo K.,Unirio | Cozman F.G.,University of Sao Paulo
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

Description logics have become a prominent paradigm in knowledge representation (particularly for the Semantic Web), but they typically do not include explicit representation of uncertainty. In this paper, we propose a framework for automatically learning a Probabilistic Description Logic from data. We argue that one must learn both concept definitions and probabilistic assignments. We also propose algorithms that do so and evaluate these algorithms on real data. © 2011 Springer-Verlag.


Armada M.,Unirio | Revoredo K.,Unirio | Luna J.E.O.,San Pablo Catholic University | Cozman F.G.,University of Sao Paulo
CEUR Workshop Proceedings | Year: 2013

Link prediction in a network is mostly based on information about the neighborhood topology of the nodes. Recently, the interest for hybrid link prediction approaches that combine topology information with information about the network individuals, has grown. However, considering the whole set of individuals may not be necessary and sometimes not even suitable. Therefore, mechanisms to automatically discover the relevant set of individuals are demanding. In this paper, we encompass this problem by proposing an algorithm that combines structure and semantic metrics to find the set of relevant individuals. We empirically evaluate this proposal analyzing the assertion role of these individuals when predicting a link through a probabilistic ontology.


da Silva G.A.R.,UNIRIO | Boechat T.O.,HEMORIO | Ferry F.R.A.,UNIRIO | Pinto J.F.C.,UNIRIO | And 4 more authors.
Revista do Instituto de Medicina Tropical de Sao Paulo | Year: 2014

Visceral leishmaniasis is an anthropozoonosis that is caused by protozoa of the genus Leishmania, especially Leishmania (Leishmania) infantum, and is transmitted to humans by the bite of sandflies of the genus Lutzomyia, such as Lutzomyia longipalpis. There are many reservoirs, including Canis familiaris. It is a chronic infectious disease with systemic involvement that is characterized by three phases: the initial period, the state period and the final period. The main symptoms are fever, malnutrition, hepatosplenomegaly, and pancytopenia. This article reports a case of a patient diagnosed with visceral leishmaniasis in the final period following autochthonous transmission in the urban area of Rio de Janeiro. The case reported here is considered by the Municipal Civil Defense and Health Surveillance of Rio de Janeiro to be the first instance of autochthonous visceral leishmaniasis in humans in the urban area of this city. The patient was discharged and is undergoing a follow-up at the outpatient clinic, demonstrating clinical improvement.


Signorini D.J.H.P.,Rio de Janeiro State Federal University | Monteiro M.C.M.,Hospital Universitario Gaffree e Guinle | de Andrade M.F.C.,Federal University of Rio de Janeiro | Signorini D.H.,Estácio de Sá University | Eyer-Silva W.A.,UNIRIO
Revista da Associacao Medica Brasileira | Year: 2012

Objective: Prevalence of chronic complications of HIV infection is increasing and early recognition and treatment of the components of metabolic syndrome (MS) are essential to prevent cardiovascular and metabolic complications. Considering this, we performed a cross-sectional study on the prevalence and risk-factors for MS among HIV-infected subjects. Methods: A total of 819 patients followed at a large outpatient HIV unit were assessed by an interviewer-administered questionnaire that recorded several demographic, epidemiologic, clinical, laboratory, and social variables. Lipodystrophy diagnosis relied on agreement between patient's self-report and physician's observation of altered body-fat deposits. The presence of three or more of the following characteristics identified MS: increased waist circumference, hypertriglyceridemia, low HDL cholesterol level, hypertension, and hyperglycemia. We used logistic regression analyses to study variables independently associated with MS. Results: The prevalence of MS was 20.6% and that of lipodystrophy was 38.5%. 61 (36.1%) out of 169 patients with MS had also lipodystrophy. Patients with metabolic syndrome were significantly more likely to be older (OR = 1.08), had higher CD4 counts (OR = 1.001), had an increased body mass index (OR = 1.27) and had longer exposure to antiretroviral therapy (OR = 1.01) than those without metabolic syndrome. Conclusion: Both traditional risk factors for cardiovascular disease and factors associated with HIV infection itself, such as an increased CD4 cell count and a longer exposure to antiretroviral therapy, seem to be associated with metabolic syndrome in the present study population. © 2012 Elsevier Editora Ltda.


Turchetto-Zolet A.C.,Federal University of Rio Grande do Sul | Pinheiro F.,Institute Botanica | Salgueiro F.,UNIRIO | Palma-Silva C.,Institute Botanica
Molecular Ecology | Year: 2013

The South American continent is composed of several biogeographical regions harbouring the highest biodiversity on the globe, encompassing five of the world's biodiversity 'hot spots'. Nonetheless, the patterns and processes responsible for shaping its astonishing species diversity are largely unknown. Here, we present a review of current South American phylogeographical knowledge based on published articles on this topic. An appraisal of the literature reveals emerging phylogeographical patterns in the biota of South America. The striking phylogeographical divergence observed among organism lineages in South American studies is suggestive of high levels of undocumented species diversity. The interplay between Pleistocene climatic oscillations and Pliocene/Miocene orogenic events has contributed to shaping the current diversity and distribution of modern lineages in both the tropical and temperate regions of South America. Although older divergence times were observed for a range of species, most herpetofauna underwent an intraspecific lineage split much earlier than other organisms. The geographical ranges of species associated with forest habitats were reduced mainly during glacial cycles, whereas species associated with open vegetation domains have shown variable responses to climatic oscillations. The results suggest a highly complex mosaic of phylogeographical patterns in South America. We suggest future research directions to promote a better understanding of the origin and maintenance of the South American biota. © 2012 Blackwell Publishing Ltd.


Engiel P.,Pontifical Catholic University of Rio de Janeiro | Cappelli C.,UNIRIO | Do Prado Leite J.C.S.,Pontifical Catholic University of Rio de Janeiro
Proceedings of the ACM Symposium on Applied Computing | Year: 2014

Lately, organizations have been subject to regulation promoting information transparency; one example of this is the Brazilian Information Access Law. This paper presents a novel way of performing requirements elicitation using both the law and a Non Functional Requirements Patterns catalog as the information sources. Since organizations must follow the law, its information systems must also implement the law as requirements. Our process is guided by pattern matching, text mining and grounded analysis. We examine the special case of the Brazilian Access Law using our approach, which compares a previously encoded transparency knowledge base with the law. Copyright 2014 ACM.


Nogueira Santos F.J.,UNIRIO | Sampaio Do Prado Leite J.C.,Pontifical Catholic University of Rio de Janeiro | Cappelli C.,UNIRIO | Batista T.V.,Federal University of Rio Grande do Norte | Santoro F.M.,UNIRIO
Proceedings of the 2011 International Workshop on Early Aspects, EA'11 | Year: 2011

Modularization of business process models using aspects is a new line of research. Aspectual languages in this context have been proposed and are still under development. However, most of the guidelines for aspects identification in this domain is still dependent on the idea of process model elements repetition. In this paper, we explore a new way of identifying aspects, based on the ideas of the GPM (Generic Process Model) and process goals. We present a procedure for aspect identification and provide an example to illustrate its application.


Araujo E.T.,UNIRIO | Ciarlini A.E.M.,UNIRIO
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

This paper proposes a language based on temporal logics to specify constraints on nondeterministic stories, generated by interactive storytelling systems, taking into account that time is continuous and branched. These constraints are checked, in real time, during story generation. Constraints to be checked are described by temporal formulas over the properties of the story. To make possible the verification in real time, it is assumed that these properties are piecewise-linear functions. A prototype, using Constraint Logic Programming for the verification of properties, is operational. © 2011 Springer-Verlag.


Guilherme Da Silva F.A.,UNIRIO | Ciarlini A.E.M.,UNIRIO | Siqueira S.W.M.,UNIRIO
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2010

Interactive storytelling is a new kind of digital entertainment in which users interact with a story being told. In this context, automated planning algorithms are interesting alternatives to create plots that are coherent with the intended genre. Diversity of stories and opportunities for interaction are however other key requirements to be considered by such applications. It is also important to generate stories incorporating failed attempts to achieve goals, since this is essential to create dramatic tension. In this paper, we describe a planner that takes these issues into account by means of planning with nondeterministic events, that is, events with more than one possible different outcome. The implemented planner has been incorporated to the new version of the interactive storytelling system Logtell. © 2010 Springer-Verlag.

Loading UNIRIO collaborators
Loading UNIRIO collaborators