No description available.

Thesis (Ph.D.) -- University of Adelaide, Dept. of Psychology, 1999. / Bibliography: leaves 259-272.

Information extraction (IE) is the process of automatically transforming written natural language (i.e., text) into structured information, such as a knowledge base. However, because natural language is inherently ambiguous, this transformation process is highly complex. On the other hand, as Information Extraction moves from the analysis of scientific documents to the analysis of Internet textual content, we cannot rely completely on the assumption that the content of the text is correct. Indeed, in contrast to scientific documents, which are peer reviewed, Internet content is not verified for the quality and correctness. Thus, two main issues that affect the IE process are the complexity of the extraction process and the quality of the data. In this dissertation, we propose an improved ontology-based IE (OBIE) by providing solutions to these issues of accuracy and content quality. Based on a hybrid strategy that combines aspects of IE that are usually considered as opposite to each other, or that are not even considered, we intend to improve IE by developing a more accurate extraction and new functionality (semantic error detection). Our approach is based on OBIE, a sub-area of IE, which reduces extraction complexity by including domain knowledge, in the form of concepts and relationships of the domain, to guide the extraction process. We address the complexity of extraction by combining information extractors that have different implementations. By integrating different types of implementation into one extraction system, we can produce a more accurate extraction. For each concept or relationship in the ontology, we can select the best implementation for extraction, or we can combine both implementations under an ensemble learning schema. In tandem, we address the quality of information by determining its semantic correctness with regard to domain knowledge. We define two methods for semantic error detection: by predefining the types of errors expected in the text or by applying logic reasoning to the text. This dissertation includes both published and unpublished coauthored material.

Thesis (Ph. D.)--University of Washington, 1985. / Vita. Bibliography: leaves [131]-137.

Thesis (M.A.)--University of Toronto, 2004. / Adviser: Janet Astington.

Equational reasoning with string diagrams provides an intuitive method for proving equations between morphisms in various forms of monoidal category. !-Graphs were introduced with the intention of reasoning with infinite families of string diagrams by allowing repetition of sub-diagrams. However, their combinatoric nature only allows commutative nodes. The aim of this thesis is to extend the !-graph formalism to remove the restriction of commutativity and replace the notion of equational reasoning with a natural deduction system based on first order logic. The first major contribution is the syntactic !-tensor formalism, which enriches Penrose's abstract tensor notation to allow repeated structure via !-boxes. This will allow us to work with many noncommutative theories such as bialgebras, Frobenius algebras, and Hopf algebras, which have applications in quantum information theory. A more subtle consequence of switching to !-tensors is the ability to definitionally extend a theory. We will demonstrate how noncommutativity allows us to define nodes which encapsulate entire diagrams, without inherently assuming the diagram is commutative. This is particularly useful for recursively defining arbitrary arity nodes from fixed arity nodes. For example, we can construct a !-tensor node representing the family of left associated trees of multiplications in a monoid. The ability to recursively define nodes goes hand in hand with proof by induction. This leads to the second major contribution of this thesis, which is !-Logic (!L). We extend previous attempts at equational reasoning to a fully fledged natural deduction system based on positive intuitionistic first order logic, with conjunction, implication, and universal quantification over !-boxes. The key component of !L is the principle of !-box induction. We demonstrate its application by proving how we can transition from fixed to arbitrary arity theories for monoids, antihomomorphisms, bialgebras, and various forms of Frobenius algebras. We also define a semantics for !L, which we use to prove its soundness. Finally, we reintroduce commutativity as an optional property of a morphism, along with another property called symmetry, which describes morphisms which are not affected by cyclic permutations of their edges. Implementing these notions in the !-tensor language allows us to more easily describe theories involving symmetric or commutative morphisms, which we then demonstrate for recursively defined Frobenius algebra nodes.

Thesis (All-College Honors) - - State University of New York College at Cortland, 2006 - - Department of Philosophy.

Thesis (Ph.D.)--Baylor University, 2008. / Bibliographic references (p. 157-158)

Quando um robô autônomo tenta resolver as tarefas de navegação dentro de um ambiente real interno usando relações qualitativas, vários problemas aparecem tais como observação parcial do ambiente e percepção incerta. Isso ocorre porque os sensores do robô não proporcionam informação suficiente para perceber completamente as situações do ambiente, além de incorporarem ruído no processo. A web semântica dota o robô autônomo com a habilidade de obter conhecimento de senso comum extraído da web, conhecimento este que os sensores do robô não podem proporcionar. Porém, nem sempre é fácil levar efetivamente estes recursos semânticos da web ao uso prático. Neste trabalho, foi examinado o uso de recursos semânticos da web na navegação robótica; mais especificamente, em uma navegação qualitativa onde o raciocínio incerto desempenha um papel significativo. Nós avaliamos o uso de uma representação relacional; particularmente, na combinação da informação semântica web e dos dados de baixo nível proporcionados pelos sensores, permitindo uma descrição de objetos e das relações entre os mesmos. Esta representação também permite o uso de abstração e generalização das situações do ambiente. Este trabalho propõe a arquitetura Web-based Relational Robotic Architecture (WRRA )para navegação robótica que combina os dados de baixo nível dos sensores do robô e os recursos web semânticos existentes baseados em lógica descritiva probabilística, como aprendizagem e planejamento relacional probabilístico. Neste trabalho, mostramos os benefícios desta arquitetura em um robô simulado, apresentando um estudo de caso sobre como os recursos semânticos podem ser usados para lidar com a incerteza da localização e o mapeamento em um problema prático. / When an autonomous robot attempts to solve navigation tasks in a qualitative relational way within a real indoor environments, several problems appear such as partial observation of the environment, and uncertain perception, since the robots sensors do not provide enough information to perceive completely the environment situations, besides the sensors incorporate noise in the process. The semantic web information endows the autonomous robot with the ability to obtain common sense knowledge from the web that the robot\'s sensors cannot provide. However, it is not always easy to effectively bring these semantic web resources into practical use. In this work, we examine the use of semantic web resources in robot navigation; more specifically, in qualitative navigation where uncertain reasoning plays a significant role. We evaluate the use of a relational representation; particularly, in the combination of the semantic web and the low-level data sensor information, which allows a description of relationships among objects. This representation also allows the use of abstraction and generalization of the environment situations. This work proposes the framework Web-based Relational Robotic Architecture WRRA for robot navigation that connects the low-level data from robot\'s sensors and existing semantic web resources based on probabilistic description logics, with probabilistic relational learning and planning. We show the benefits of this framework in a simulated robot, presenting a case study on how semantic web resources can be used to face location and mapping uncertain in a practical problem.

Quando um robô autônomo tenta resolver as tarefas de navegação dentro de um ambiente real interno usando relações qualitativas, vários problemas aparecem tais como observação parcial do ambiente e percepção incerta. Isso ocorre porque os sensores do robô não proporcionam informação suficiente para perceber completamente as situações do ambiente, além de incorporarem ruído no processo. A web semântica dota o robô autônomo com a habilidade de obter conhecimento de senso comum extraído da web, conhecimento este que os sensores do robô não podem proporcionar. Porém, nem sempre é fácil levar efetivamente estes recursos semânticos da web ao uso prático. Neste trabalho, foi examinado o uso de recursos semânticos da web na navegação robótica; mais especificamente, em uma navegação qualitativa onde o raciocínio incerto desempenha um papel significativo. Nós avaliamos o uso de uma representação relacional; particularmente, na combinação da informação semântica web e dos dados de baixo nível proporcionados pelos sensores, permitindo uma descrição de objetos e das relações entre os mesmos. Esta representação também permite o uso de abstração e generalização das situações do ambiente. Este trabalho propõe a arquitetura Web-based Relational Robotic Architecture (WRRA )para navegação robótica que combina os dados de baixo nível dos sensores do robô e os recursos web semânticos existentes baseados em lógica descritiva probabilística, como aprendizagem e planejamento relacional probabilístico. Neste trabalho, mostramos os benefícios desta arquitetura em um robô simulado, apresentando um estudo de caso sobre como os recursos semânticos podem ser usados para lidar com a incerteza da localização e o mapeamento em um problema prático. / When an autonomous robot attempts to solve navigation tasks in a qualitative relational way within a real indoor environments, several problems appear such as partial observation of the environment, and uncertain perception, since the robots sensors do not provide enough information to perceive completely the environment situations, besides the sensors incorporate noise in the process. The semantic web information endows the autonomous robot with the ability to obtain common sense knowledge from the web that the robot\'s sensors cannot provide. However, it is not always easy to effectively bring these semantic web resources into practical use. In this work, we examine the use of semantic web resources in robot navigation; more specifically, in qualitative navigation where uncertain reasoning plays a significant role. We evaluate the use of a relational representation; particularly, in the combination of the semantic web and the low-level data sensor information, which allows a description of relationships among objects. This representation also allows the use of abstraction and generalization of the environment situations. This work proposes the framework Web-based Relational Robotic Architecture WRRA for robot navigation that connects the low-level data from robot\'s sensors and existing semantic web resources based on probabilistic description logics, with probabilistic relational learning and planning. We show the benefits of this framework in a simulated robot, presenting a case study on how semantic web resources can be used to face location and mapping uncertain in a practical problem.