The Use of Latent Semantic Indexing to Mitigate OCR Effects of Related Document Images

Due to both the widespread and multipurpose use of document images and the current availability of a high number of document images repositories, robust information retrieval mechanisms and systems have been increasingly demanded. This paper presents an approach to support the automatic generation of relationships among document images by exploiting Latent Semantic Indexing (LSI) and Optical Character Recognition (OCR). We developed the LinkDI (Linking of Document Images) service, which extracts and indexes document images content, computes its latent semantics, and defines relationships among images as hyperlinks. LinkDI was experimented with document images repositories, and its performance was evaluated by comparing the quality of the relationships created among textual documents as well as among their respective document images. Considering those same document images, we ran further experiments in order to compare the performance of LinkDI when it exploits or not the LSI technique. Experimental results showed that LSI can mitigate the effects of usual OCR misrecognition, which reinforces the feasibility of LinkDI relating OCR output with high degradation.

UQ eSpace Users - Frequently Asked Questions

A FAQ on how to use the UQ eSpace repository.

UQ eSpace Poster

UQ eSpace (http://espace.uq.edu.au/) is The University of Queensland's institutional digital repository. The poster outlines all the different ways academic staff and postgraduate students can make use of the repository.

GIS/GPS to document increased access to community-based treatment for tuberculosis in Africa

Application of geographic information system (GIS) and global positioning system (GPS) technology in the Hlabisa community-based tuberculosis treatment programme documents the increase in accessibility to treatment after the expansion of the service from health facilities to include community workers and volunteers.

Document retrieval for e-mail search and discovery using Formal Concept Analysis

This paper discusses a document discovery tool based on Conceptual Clustering by Formal Concept Analysis. The program allows users to navigate e-mail using a visual lattice metaphor rather than a tree. It implements a virtual. le structure over e-mail where files and entire directories can appear in multiple positions. The content and shape of the lattice formed by the conceptual ontology can assist in e-mail discovery. The system described provides more flexibility in retrieving stored e-mails than what is normally available in e-mail clients. The paper discusses how conceptual ontologies can leverage traditional document retrieval systems and aid knowledge discovery in document collections.

On the specification of a component repository

The lack of a commonly accepted de nition of a software component, the proliferation of competing `standards' and component frameworks, is here to stay, raising the fundamental question in component-based development of how to cope in practice with heterogeneity. This paper reports on the design of a Component Repository aimed to give at least a partial answer to the above question. The repository was fully speci ed in Vdm and a working prototype is currently being used in an industrial environment

Solução doméstica para televisão digital

Mestrado em Engenharia Electrotécnica e de Computadores

Visualização com nuvens de tags

Com a evolução da Internet houve uma consequente transformação da forma como os resultados são apresentados e das técnicas de visualização e apresentação de conteúdos. A presente dissertação é sobre visualização com nuvens de tags e foi a conjugação de diversos factores que originou a investigação sobre este tema. Uma nuvem de tags (em inglês: tag clouds) é uma concepção visual de um conjunto de tags isoladas com uma representação ilustrativa da sua importância, ou seja visualmente hierarquizadas e em que cada tag se hiperliga a um ou mais sítios Web ou documentos. São extremamente úteis para a navegação ou para a descoberta de informação genérica, podendo requerer menor carga cognitiva do utilizador durante consultas e pesquisas se alguns aspectos forem considerados na sua construção. Nesta dissertação discutem-se alguns dos factores visuais que podem contribuir para que as nuvens de tags sejam ferramentas efectivas e intuitivas para os utilizadores. O conceito de folksonomia está relacionado com nuvens de tags, permitindo categorizar as definições atribuídas a conteúdos recorrendo a palavras-chave (tags). Neste documento são ainda exploradas as suas vantagens e desvantagens. Discutem-se ainda alguns modos de visualização e apresentação de nuvens de tags, desde o tipo de ordenação ao algoritmo de geração da nuvem de tags. Aborda-se ainda o impacto do tipo de fonte e a utilização de cores monocromáticas ou coloridas que estas possam ter na visualização da nuvem. A própria tag também é categorizada e explorada ao nível do seu posicionamento na nuvem e a percepção que causa no utilizador. Sobre a visualização de nuvens de tags, refere-se ainda que as conclusões obtidas foram utilizadas num repositório de objectos educativos.

CrimsonHex: a learning objects repository for programming exercises

A repository of learning objects is a system that stores electronic resources in a technology-mediated learning process. The need for this kind of repository is growing as more educators become eager to use digital educa- tional contents and more of it becomes available. The sharing and use of these resources relies on the use of content and communication standards as a means to describe and exchange educational resources, commonly known as learning objects. This paper presents the design and implementation of a service-oriented reposi- tory of learning objects called crimsonHex. This repository supports new definitions of learning objects for specialized domains and we illustrate this feature with the definition of programming exercises as learning objects and its validation by the repository. The repository is also fully compliant with existing commu- nication standards and we propose extensions by adding new functions, formalizing message interchange and providing a REST interface. To validate the interoperability features of the repository, we developed a repository plug-in for Moodle that is expected to be included in the next release of this popular learning management system.

Extensible metadata repository for information systems

Thesis submitted to Faculdade de Ciências e Tecnologia of the Universidade Nova de Lisboa, in partial fulfillment of the requirements for the degree of Master in Computer Science

CrimsonHex: a service oriented repository of specialised learning objects

The corner stone of the interoperability of eLearning systems is the standard definition of learning objects. Nevertheless, for some domains this standard is insufficient to fully describe all the assets, especially when they are used as input for other eLearning services. On the other hand, a standard definition of learning objects in not enough to ensure interoperability among eLearning systems; they must also use a standard API to exchange learning objects. This paper presents the design and implementation of a service oriented repository of learning objects called crimsonHex. This repository is fully compliant with the existing interoperability standards and supports new definitions of learning objects for specialized domains. We illustrate this feature with the definition of programming problems as learning objects and its validation by the repository. This repository is also prepared to store usage data on learning objects to tailor the presentation order and adapt it to learner profiles.