Over half of the world's population are at risk of infection from dengue fever (Guha-Sapir 2005). This viral disease is transmitted by the female Aedes aegypti mosquito and is the major source of human death in the world when compared with any other vector borne disease (Gubler1998a). The first important epidemic of dengue haemorrhagic fever (DHF) in America was reported in Cuba in 1981 and subsequently in Venezuela during 1989 and 1990 (Oletta2006, Brightmer1998). There has been a trend of increased incidence in many Central and South American countries since 1990 - Brazil, Venezuela, Honduras and Mexico (SanMartin2010) with Venezuela having the highest number of cases of DHF. The urgent need for more effective public health measures to combat this disease in Venezuela drove the decision to undertake the work described in this dissertation. Spatio-Temporal modelling has been developed for the prediction of the occurrence of dengue fever in Zulia state, Venezuela. A systematic approach has been adopted to validate this tool. At the first stage of the analysis an exploratory study was performed to underline the most significant features of the dynamics of incidence rates of dengue fever from 2002 to 2008. In the second stage a Generalized Linear Model (GLM) approach was used in the form of Negative Binomial Generalized Linear Mixed model (GLMM) to compare Relative Risk (RR) across exposure groups by age and sex, using an epidemiological dataset covering the whole of Zulia State, Venezuela. This approach used both a frequentist and a Bayesian perspective for comparative purposes of both outcomes and methodologies. Finally a Spatio-Temporal model was constructed based on Generalized Additive Mixed model (GAMM) framework because the earlier analysis identified a complex association between covariates and response variables. This GAMM structure was further developed so that it could be used to help predict future outbreaks of the disease in Zulia state with a good degree of accuracy.

Master of Science / Department of Statistics / Michael J. Higgins / There have been a wave of recent interest in understanding the dynamics of vehicular emissions in university towns. Using data from a recent survey of Kansas State University students, faculty, and staff-which includes a detailed itinerary of a day’s worth of travel -I assess spatial and temporal trends of emissions at Kansas State University. By combining the survey data with secondary sources of data, including vehicular emissions data from Fueleconony.gov and trip distance information from Google, I obtain improved measure for the quantity of emissions produced during each trip. After an extensive cleaning of the data, I develop heat maps for emissions composed of shape files identifying zip-code boundaries and a raster layer. I find that most vehicular emissions are concentrated around campus, with the highest level of emissions occurring during rush hour. Furthermore, faculty and staff appear to, on average, produce more emissions than students. I also investigate how proposed methods for reducing emissions will affect these spatio-temporal trends. Specifically, I show that walking or biking short distances instead of driving may lead to a small overall reduction in vehicular emissions.

O fator temporal é uma variável natural da maioria dos sistemas de informação, pois no mundo real os eventos ocorrem de maneira dinâmica, modicando continuamente os valores dos seus objetos no decorrer do tempo. Muitos desses sistemas precisam registrar essa modicação e atribuir os instantes de tempo em que cada informação foi válida no sistema. Este trabalho reúne as características relacionadas aos Bancos de Dados Temporais e Bancos de Dados Objeto-Relacionais. O objetivo primordial é propor uma forma de implementar alguns aspectos temporais, desenvolvendo um módulo que faça parte das características e funcionalidades internas de um SGBD. O módulo temporal contempla principalmente a parte de restrições de integridade temporal que é utilizada para manter a consistência da informação temporal armazenada. Para isso, é proposto um novo tipo de dado que melhor representa as marcas temporais dos objetos. Uma parte importante para a implementação desse projeto é a utilização de um SGBD objeto-relacional que possui algumas características orientadas a objetos que permitem a extensão de seus recursos, tornando-o capaz de gerenciar alguns aspectos temporais. O módulo temporal desenvolvido torna esses aspectos temporais transparentes para o usuário. Por conseqüência, esses usuários são capazes de utilizar os recursos temporais com maior naturalidade. / The temporal factor is a natural variable of the majority of the information systems, therefore in the real world the events occur in dynamic way, modifying continuously the values of its objects in elapsing of the time. Many of these systems need to register this modication and to attribute the instants of time where each information was valid in the system. This work congregates the characteristics related to the Temporal Databases and Object-Relational Databases. The primordial objective is to consider a form to implement some temporal aspects, developing a module that is part of the characteristics and internal functionalities of a DBMS. The temporal module mainly contemplates the part of restrictions of temporal integrity that is used to keep the consistency of the stored temporal information. For this, a new data type is proposed that better represent the objects timestamps. An important part for the implementation of this project is the use of a object-relational DBMS that has some object-oriented characteristics that allow the extension of its resources, becoming capable to manage some temporal aspects. The developed temporal module becomes these transparent temporal aspects for the user. For consequence, these users are capable to use the temporal resources more naturally.

No description available.

Thesis (Ph. D.)--State University of New York at Binghamton, Anthropology Dept., 2006. / Includes bibliographical references.

The underlying assumptions for interpreting the meaning of data often change over time, which further complicates the problem of semantic heterogeneities among autonomous data sources. As an extension to the COntext INterchange (COIN) framework, this paper introduces the notion of temporal context as a formalization of the problem. We represent temporal context as a multi-valued method in F-Logic; however, only one value is valid at any point in time, the determination of which is constrained by temporal relations. This representation is then mapped to an abductive constraint logic programming framework with temporal relations being treated as constraints. A mediation engine that implements the framework automatically detects and reconciles semantic differences at different times. We articulate that this extended COIN framework is suitable for reasoning on the Semantic Web. / Singapore-MIT Alliance (SMA)

The underlying assumptions for interpreting the meaning of data often change over time, which further complicates the problem of semantic heterogeneities among autonomous data sources. As an extension to the COntext INterchange (COIN) framework, this paper introduces the notion of temporal context as a formalization of the problem. We represent temporal context as a multi-valued method in F-Logic; however, only one value is valid at any point in time, the determination of which is constrained by temporal relations. This representation is then mapped to an abductive constraint logic programming framework with temporal relations being treated as constraints. A mediation engine that implements the framework automatically detects and reconciles semantic differences at different times. We articulate that this extended COIN framework is suitable for reasoning on the Semantic Web. / Singapore-MIT Alliance (SMA)

The underlying assumptions for interpreting the meaning of data often change over time, which further complicates the problem of semantic heterogeneities among autonomous data sources. As an extension to the COntext INterchange (COIN) framework, this paper introduces the notion of temporal context as a formalization of the problem. We represent temporal context as a multi-valued method in F-Logic; however, only one value is valid at any point in time, the determination of which is constrained by temporal relations. This representation is then mapped to an abductive constraint logic programming framework with temporal relations being treated as constraints. A mediation engine that implements the framework automatically detects and reconciles semantic differences at different times. We articulate that this extended COIN framework is suitable for reasoning on the Semantic Web. / Singapore-MIT Alliance (SMA)

The underlying assumptions for interpreting the meaning of data often change over time, which further complicates the problem of semantic heterogeneities among autonomous data sources. As an extension to the COntext INterchange (COIN) framework, this paper introduces the notion of temporal context as a formalization of the problem. We represent temporal context as a multi-valued method in F-Logic; however, only one value is valid at any point in time, the determination of which is constrained by temporal relations. This representation is then mapped to an abductive constraint logic programming framework with temporal relations being treated as constraints. A mediation engine that implements the framework automatically detects and reconciles semantic differences at different times. We articulate that this extended COIN framework is suitable for reasoning on the Semantic Web. / Singapore-MIT Alliance (SMA)

This report deals with the issue of tense and aspect system in Korean both semantically and syntactically. Previous accounts for tense and aspect in Korean have been based on the assumption that the suffix has a fixed meaning, either temporally or aspectually, and either perfect(ive) or imperfective, and so on. In this report, I argue that the interpretation of the temporal suffix in Korean must be more flexible. The interpretation of the suffix is dependent on the occurrence conditions such as the characteristics of the verbs (or the predicates) and the temporal adverbials. For this claim, I argue how to understand tense and viewpoint aspect in Korean based on ‘twocomponent theory of aspect’ (Smith 1997) along the line of ‘boundedness’ of the grammaticalized aspect (Depraetere 1995) and ‘Event realization’ by Bohnemeyer and Swift (2004): –ess is either past tense or perfective aspect suffix and –nun is a present tense suffix with imperfective aspect meaning, because –nun only occurs with [+dynamic] predicates. Further, I argue how we can reflect the semantic tense and aspect onto the syntactic representation properly. Following Demirdache & Uribe-Etxebarria (1997, 2000), I show how tense and aspect in Korean can be mapped onto the syntactic representation. To treat the temporal adverbials, I follow Thompson (2005) that temporal adverbials are linked to [spec, AspP] position when they play a role as Reference Time (RT). Finally, I argue how the relations of the temporal arguments are syntactically realized in Korean. And –ess is interpreted as perfective aspect when the RT and ST (Speech Time) overlap. / text