GeoMiningVisualQL: uma linguagem de consulta visual para mineração de dados geográficos

• Ano de defesa: 2010
• Autor(a) principal: Pedrosa, Klebber de Araújo
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraí­ba; BR; Informática; Programa de Pós Graduação em Informática; UFPB
• Palavras-chave em Português: Bancos de Dados Geográficos; Descoberta de Conhecimento; Linguagem de Consulta Visual; Geographic Databases; Knowledge Discovery; Visual Query Language; CNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/6146

Several areas of knowledge domain, such as remote sensing systems, transportation, telecommunication, digital mapping, among others, make use of large amounts of geographic data. Typically, these data are stored in Management Systems Geographic Database (SGBDGeo), through which can be often manipulated by Geographic Information Systems (GIS). However, these systems are not able to extract new information, previously unknown to users, which may be embedded within the database field analysed and that, somehow, represent new and userful knowledge, for example, for decision making. In this case, it is necessary to make use of specific techniques of Knowledge Discovery in Databases (KDD). Moreover, spatial data present inherently visual characteristics that, often, can be associated with geometric and pictographic visual representations. In this context, there are few visual query languages for spatial data. However, few of this treat mining methods among the spatial data. Thus, this paper proposes the construction of an environment for data mining tasks performed under certain geographical areas, beyond the formal specification of a visual query language to be used in this environment. These queries are formulated through pictorial representations of geographic features, operators, and spatial relationships between these data. To this end, we use metaphorical abstractions on the metadata of the geographical environment, and the approach defined as "flowing stream" in which the user focuses attention on certain stages of the mining process, facilitating the construction of these consultations a number of them. Thus, the proposed environment aims to simplify the tasks of consultations on mining spatial data, making them more user friendly, providing more efficiency and speed when compared to textual queries scripts.