MINERAÇÃO DE DADOS: ALGORITMO DA CONFIANÇA INVERSA

Detalhes bibliográficos
Ano de defesa: 2005
Autor(a) principal: Casanova, Anderson Araújo
Orientador(a): LABIDI, Sofiane lattes
Banca de defesa: Não Informado pela instituição
Tipo de documento: Dissertação
Tipo de acesso: Acesso aberto
Idioma: por
Instituição de defesa: Universidade Federal do Maranhão
Programa de Pós-Graduação: PROGRAMA DE PÓS-GRADUAÇÃO EM ENGENHARIA DE ELETRICIDADE/CCET
Departamento: Engenharia
País: BR
Palavras-chave em Português:
Palavras-chave em Inglês:
Área do conhecimento CNPq:
Link de acesso: http://tedebc.ufma.br:8080/jspui/handle/tede/373
Resumo: This work presents studies that culminated in the development of a data mining algorithm that extracts knowledge in a more efficient way and allows for a better use of the collected information. Decisions based on imprecise information and a lack of criteria can cause the relatively few resources available to be poorly applied, burdening taxpayers and consequently the state. This much-needed information which allows for the fairest and most efficient application of available resources and which would facilitate the work of the users as well as those who render the services should be based upon consideration of the great variety of established criteria. The making of a decision should be based upon the evaluation of the most varied types of data and be analyzed by specialists who can judge which are true needs, so that the criteria for the search of knowledge may be defined. The Algorithm of Inverse Confidence - ACI accomplishes data mining using the technique of association rules, and it proposes a new measure that enlarges the dimension of extracted information through five fixed rules. ACI also classifies and associates items, using the concept of the fuzzy logic, through parameters established by the user. ACI was applied in the surgical center of HUUFMA - Academical Hospital of the Federal University of Maranhão - envisioning the extraction of knowledge (standards).