Analysis of matrix reordering algorithms based on multiple binarization

• Grant number: 15/00411-6
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): April 01, 2015
• Effective date (End): March 31, 2016
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Celmar Guimarães da Silva
• Grantee: Maressa Rodrigues da Silva
• Host Institution: Faculdade de Tecnologia (FT). Universidade Estadual de Campinas (UNICAMP). Limeira , SP, Brazil

Abstract

Matrices are data structures that underlies many types of visualizations, such as heatmaps. Lots of algorithms permute their rows and columns in order to improve visual understanding and highlight patterns. Previous works used Matrix Reordering Analyzer (MRA) tool for helping to create new reordering algorithms (MB and SMB, based on multiple binarization of matrices) and specially for comparing reordering quality and execution time of algorithms. Analysis of these algorithms pointed out good results for reordering matrices whose data obey some canonical data pattern defined by the literature. However, algorithms SMB and MB were tested only with canonical patterns configured with fixed parameters. Therefore, this research project aims to increase the analysis of applying algorithms SMB and MB for reordering matrices created from canonical data patterns, taking into consideration possible variations of these patterns. Besides, this work also aims to adapt MRA for automatic creation of graphics about reordering experiments, which currently is made through spreadsheet editors.

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

DA SILVA, CELMAR GUIMARAES; MEDINA, BRUNO FIGUEIREDO; DA SILVA, MARESSA RODRIGUES; KAWAKAMI, WILLIAN HITOSHI; NAVES ROCHA, MIGUEL MECHI. A fast feature vector approach for revealing simplex and equi-correlation data patterns in reorderable matrices. INFORMATION VISUALIZATION, v. 16, n. 4, p. 261-274, OCT 2017. (15/14854-7, 14/11186-0, 15/00411-6)