Processo de descoberta de conhecimento em dados não-estruturados : estudo de caso para a inteligência competitiva

AUTOR(ES)

José Marcelo Pereira de Araújo

DATA DE PUBLICAÇÃO

2007

RESUMO

Processes of knowledge discovery in data not structured obtained from books, documents and electronic reports can provide the acquisition of information to subside processes of decisions and actions of Competitive Intelligence. To discover knowledge in data not structured, it is necessary to obtain the dominant concepts of information, that is, their main ideas to classify them in groups. This classification derives from the occurrence of vocabularies contained in information. The discovery process of knowledge in data together with a semantics intelligence classifier can serve as a facilitator in the structuration, analysis, grouping and automatic classification of information constituted in texts. The Competitive Intelligence characterizes as a potential kind of user of such technology, by reason of the bulk of produced texts and transmitted in several digital media since it demands velocity and accuracy in the semantics analysis of the produced documentation. This dissertation proposes a discovery process of the knowledge in data not structured, using the neural artificial Web ART 1 as a semantics classifier. The aim is to raise to a power the area of Competitive Intelligence with a classification of texts according to the source sample SWOT, characterizing by the classification of texts in the quadrants SWOT that involve weakness, fortress, advantages and disadvantages about an institution. Besides, the waited outcomes involve a processual methodology appropriate to the development of a discovery a process of knowledge that is supported process in the identification of wisdom theories and bases of support to the processes of data mining. Thus, results are waited in order to justify the main objectives of this research.

ASSUNTO(S)

text mining gestão do conhecimento; exploração de dados; tecnologia da informação; inteligência competitiva ciencia da computacao text mining rede neural art1 inteligência competitiva kdd neural art 1 web competitive intelligence kdd

Documentos Relacionados

• Uma Metodologia para Mineração de Regras de Associação Usando Ontologias para Integração de Dados Estruturados e Não-Estruturados
• Inteligência de fontes abertas : um estudo de caso sobre descoberta de conhecimento no Diário Oficial da União
• Pré-processamento no processo de descoberta de conhecimento em banco de dados
• Utilizando o processo de descoberta de conhecimento em banco de dados para identificar candidatos a padrão de análise para bancos de dados geográficos
• Ferramenta de suporte para inteligência competitiva: um estudo de caso na Embrapa