Integrating distance functions and feature extractors into DBMS to perform similarity queries

• Grant number: 11/05301-3
• Support Opportunities: Scholarships in Brazil - Master
• Start date: August 01, 2011
• End date: July 31, 2013
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Caetano Traina Junior
• Grantee: Marcos Vinicius Naves Bêdo
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

The data being collected and generated nowadays increases not only in volume, but also in complexity, leading to the need of new query operators. Similarity queries are one of the most pursued resources to retrieve complex data. The most studied operators to perform similarity are the Range Query and the k-Nearest Neighbor Query. Until recently, those queries were not available in the Database Management Systems. In fact SQL, and the standard language to query relational DBMS does not have commands for similarity queries.Now, several algorithms and operators have been developed for similarity queries, so they must be integrated to existing production DBMSs. The GBdI-ICMC-USP laboratory has been developing the initial prototype of a query interpreter for an SQL extension that embodies new commands to express and to execute similarity conditions seamlessly integrated to the existing conditions of the relational environments.This project aims at continuing this development, creating the new techniques required to fully integrate similarity queries into the operational infrastructure of a relational DBMS, and in special the ability to define, manipulate and execute distance functions -- employed to evaluate the similarity between complex element pairs -- and feature extractors -- algorithms tailored to each complex data, as images, that extract the main features of each complex element that enable comparing them.The large amount of distinct distance functions and feature extractors that can be developed leads this project to create a standard to define and integrate code that implements those concepts into a DBMS. We will use large data sets of medical images and of climate time series as case studies to validate our work. (AU)

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

BÊDO, Marcos Vinicius Naves. Including distance functions and features extractors to support similarity queries. 2013. Master's Dissertation - Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB) São Carlos.