$$\newcommand{\bra}[1]{\left<#1\right|}\newcommand{\ket}[1]{\left|#1\right>}\newcommand{\bk}[2]{\left<#1\middle|#2\right>}\newcommand{\bke}[3]{\left<#1\middle|#2\middle|#3\right>}$$
X
INFORMAÇÕES SOBRE DIREITOS AUTORAIS


As obras disponibilizadas nesta Biblioteca Digital foram publicadas sob expressa autorização dos respectivos autores, em conformidade com a Lei 9610/98.

A consulta aos textos, permitida por seus respectivos autores, é livre, bem como a impressão de trechos ou de um exemplar completo exclusivamente para uso próprio. Não são permitidas a impressão e a reprodução de obras completas com qualquer outra finalidade que não o uso próprio de quem imprime.

A reprodução de pequenos trechos, na forma de citações em trabalhos de terceiros que não o próprio autor do texto consultado,é permitida, na medida justificada para a compreeensão da citação e mediante a informação, junto à citação, do nome do autor do texto original, bem como da fonte da pesquisa.

A violação de direitos autorais é passível de sanções civis e penais.
Coleção Digital

Avançada


Formato DC |



Título: KNOWLEDGE SEARCH IN DATABASES
Autor: CIBELE LUZANA REIS
Instituição: PONTIFÍCIA UNIVERSIDADE CATÓLICA DO RIO DE JANEIRO - PUC-RIO
Colaborador(es):  EMMANUEL PISECES LOPES PASSOS - ADVISOR
Nº do Conteudo: 11103
Catalogação:  27/12/2007 Idioma(s):  PORTUGUESE - BRAZIL
Tipo:  TEXT Subtipo:  THESIS
Natureza:  SCHOLARLY PUBLICATION
Nota:  Todos os dados constantes dos documentos são de inteira responsabilidade de seus autores. Os dados utilizados nas descrições dos documentos estão em conformidade com os sistemas da administração da PUC-Rio.
Referência [pt]:  https://www.maxwell.vrac.puc-rio.br/colecao.php?strSecao=resultado&nrSeq=11103@1
Referência [en]:  https://www.maxwell.vrac.puc-rio.br/colecao.php?strSecao=resultado&nrSeq=11103@2
Referência DOI:  https://doi.org/10.17771/PUCRio.acad.11103

Resumo:
This dissertation investigates the genetic algorithms and neural networks as applications tools to find knowledge, in the form of rules, from a database. This new area, KDD (Knowledge Discovery in Database) appeared with the need of developing tools that can, in automatic and intelligent way, help the data analysis to transform great volumes of data in information and to organize these information in useful knowledge. The research here summarized is therefore, a development in the area of computational systems (development of systems) and in the area of intelligence computational (data mining, genetic algoriths, neural networks, intelligence interfaces, decision support systems and creation of knowledge bases). The thesis work was divided in five main parts: A study of the KDD process: a study of the structure of the KDD systems found in the literature; the development of KDD systems, one using genetic algorithms and the others using neural networks; the study of cases and the analysis of the performance of the developed systems. The KDD process is able to find new knowledge (patterns, tendencies, facts, probability and associations) from a certain database. Basically KDD involves eight steps, that are: problem definition, data selection, cleaning, enrichment, preprocessing, coding, data mining and the reporting containing the interpretation of the results. The Data Mining is frequently seen as the key element of the KDD process. The extraction of the knowledge, itself, happens in the Data mining, where any technique that helps extract more information out of your data is useful. In Data Mining we can make use of a heterogeneous group of techiques, for example, Statistical techniques, Visualization techniques, Neural Networks and Genetic algorithms. Therefore the studies of the KDD process included studies on data mining, machine learning, data warehouse, the KDD process and the KDD environment, formal aspects of the learning algoriths, artificial intelligence, and some applications in the real life. In several KDD systems found in the literature that were studied and analyzed, we can mention systems that uses, in the data mining step, one or more of following computation techniques to extract patterns and associations from data as: visualization techniques, query tools, statistical techniques, online analytical processing (OLAP), decision trees, association rules, neural networks and genetic algorithms. In this work two KDD systems wer developed. In each one of the developed models a visualization techniques was used, to guarantee the interaction of the system with the data analyst. And in the Data Mining step, genetic algorithms was used in one of the models, and Backpropagation Neural Networks in the other. For comparison and support effect, a system was developed using Statistical techniques. The genetic algorithm model is to find the best production rule related to a database, that answers to a specific question. And the results of the Neural Networks model is to be compared with the results of the genetic algorithm model. The application phase consisted of analyzing two different databases, one with the boys´data that lives in the street, and the other with the students´data that makes the university admission test. In the analysis of the databases it was used the KDD system here developed, with the objective to find, with genetic algorithms, or Neural Network, the best production rule, related to the databases, that answers a specific question. Two types of question. Two types of question were considered, the ones that look for characteristic of a group of data, for example, Which the boys characteristics that live in the streets? And Which the characteristics of a group of individuals that were classified but they didn´t enroll in the university? And that associates groups of data, for example, What differentiate the boys, with similar economic situation, that w

Descrição Arquivo
COVER, ACKNOWLEDGEMENTS, RESUMO, ABSTRACT, SUMMARY AND LISTS  PDF  
CHAPTER 1  PDF  
CHAPTER 2  PDF  
CHAPTER 3  PDF  
CHAPTER 4  PDF  
CHAPTER 5  PDF  
CHAPTER 6  PDF  
REFERENCES AND ANNEX  PDF  
Logo maxwell Agora você pode usar seu login do SAU no Maxwell!!
Fechar Janela



* Esqueceu a senha:
Senha SAU, clique aqui
Senha Maxwell, clique aqui