Incorporation of data cluster detection techniques in relational database management systems (DBMS)

• Grant number: 07/02497-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: August 01, 2007
• End date: December 31, 2008
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Maria Camila Nardini Barioni
• Grantee: Gabriel de Souza Fedel
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

In the last decades, a great variety of data mining techniques has been developed in order to meet the objectives of several KDD (Knowledge Discovery in Databases) applications. In the beginning, the center of interest of the researches in this field was the definition of new operations and the development of new mining algorithms. The majority of the algorithms treated the database as being a repository from which the data were extracted and inserted in memory structures before being analyzed. This fact restricted the amount of data that could be efficiently manipulated by these mining algorithms. Consequently, several researchers started to consider different strategies aiming to make the KDD process a tool for real applications that store their data in database management systems (DBMS). These researchers addressed topics related to the integration of data mining techniques and relational DBMS. That is the context where this under-graduate project is placed. Specifically, the work described in this project aims to allow the specification of queries about the results of clustering processes considering initially the algorithms based on the k-medoid method.