Aplicação da teoria dos grafos na análise de similaridade e complexidade de consultas em bancos de dados relacionais

Abstract

Information Systems constant changes and refactoring processes eventually result in design debts, one of them being related to the database management. Redundancy and complexity are often found in databases and eventually affect the overall system performance. In this study, the analysis conducted was based on graph analysis, which is an essential technique in several fields demanding data management. The investigation involved the study of relationships and connections in a real-world organization’s database, between SQL queries represented as nodes and edges in a graph structure. By analyzing the graph structure and properties, it was possible to identify important nodes, detect clusters of related data, and uncover hidden relationships and redundancy. The results indicate that 50% of the database queries had medium to high similarity in subgraphs, which allows the organization to gain valuable insights into their data, make informed decisions, and optimize their database performance.