Knowledge Discovery from Unstructured Information through Correlation and Association Techniques

• Autor(a) principal: Sérgio, Marina Carradore
• Data de Publicação: 2016
• Outros Autores: da Silva, Thales do Nascimento, Gonçalves, Alexandre Leopoldo
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Em Questão (Online)
• Texto Completo: https://seer.ufrgs.br/index.php/EmQuestao/article/view/59514
• topic: Knowledge discovery. Correlation. Association. Unstructured information. Temporality.; Descoberta de conhecimento. Correlação. Associação. Informações não estruturadas. Temporalidade.

Nowadays, technology’s c rrent stat s seeks means to support the exponential increase of information available around the Internet or in organizations, and regarding that most of said information comes in a textual form, this is a challenge to the areas of crawling, storage, retrieval and analysis of information. This article aims to provide a Knowledge Discovery model based on the temporal correlation and association from large document collections. The methodology set for this process involve descriptive and explorative researches using papers taken right from the Science Direct® database as a tool for data collection and analysis. Through this kind of information is possible to extract rules, patterns, trends, and networks, all of them being usufel to the process of making decisions within organizations in order to generate competitive advantage. Thus, the main contribution of this paper relies on the proposition of a model towards the understanding of temporal aspects, considering factual relationships (through correlations) or not (through associations) between terms in a domain.