Multilevel visual representation to assist the exploration of data sets

• Grant number: 18/17881-3
• Support Opportunities: Regular Research Grants
• Start date: November 01, 2018
• End date: January 31, 2021
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Danilo Medeiros Eler
• Grantee: Danilo Medeiros Eler
• Host Institution: Faculdade de Ciências e Tecnologia (FCT). Universidade Estadual Paulista (UNESP). Campus de Presidente Prudente. Presidente Prudente , SP, Brazil
• Associated researchers: Ives Renê Venturini Pola ; Rogério Eduardo Garcia ; Ronaldo Celso Messias Correia
• Associated scholarship(s): 18/25755-8 - Multilevel visual representation to assist the exploration of data sets, BP.TT

Abstract

Multidimensional projection techniques have been widely used to aid in the research and organization of data sets. In Visualization, these techniques can represent into 2D or 3D space the existing relationships from the feature spaces that describes data sets. One issue that affects the visual exploration of graphical representations generated by projection techniques is the overlapping of markers used to represent the data set instances. The overlap may result from the dimensionality reduction itself, the high similarity between the instances, and the limited screen space used to visualize the data set. In the literature, some overlap removal techniques were used to minimize the overlapping problem, however, since then the data set size has increased, also a larger screen space has been required to visualize the entire data set, requiring the user to navigate over the projected dataset. An alternative presented by some techniques is to perform a hierarchical exploration to reduce the space required to create the visual representation or to summarize the dataset. This research project has the objective of assisting the exploration of data sets through a multilevel representation of the result obtained from multidimensional projection techniques. In this way, the user will get an overview of the data set under analysis and will explore the parts of as he/she navigates the different levels of the graphical representation generated. At each level, overlapping removal techniques may be employed to avoid overlapping of markers. In addition, representative of each level will be presented to provide a summary of the groups presented to the user. (AU)

Scientific publications

(9)

(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)

MARCILIO-JR, WILSON E.; ELER, DANILO M.; GARCIA, ROGERIO E.; CORREIA, RONALDO C. M.; RODRIGUES, RAFAEL M. B.. Visual analytics of COVID-19 dissemination in Sao Paulo state, Brazil. JOURNAL OF BIOMEDICAL INFORMATICS, v. 117, MAY 2021. (18/17881-3)

MARCILIO-JR, WILSON E.; ELER, DANILO M.. Explaining dimensionality reduction results using Shapley values. EXPERT SYSTEMS WITH APPLICATIONS, v. 178, SEP 15 2021. (18/17881-3, 18/25755-8)

MARCILIO-JR, WILSON E.; ELER, DANILO M.; PAULOVICH, V, FERNANDO; RODRIGUES-JR, JOSE F.; ARTERO, ALMIR O.. ExplorerTree: A Focus+Context Exploration Approach for 2D Embeddings. BIG DATA RESEARCH, v. 25, JUL 15 2021. (16/11707-6, 17/17450-0, 18/17881-3, 18/25755-8)

TEIXEIRA, JAQUELINE; MARCILIO-JR, WILSON E.; ELER, DANILO M.; ARTERO, ALMIR; BRANDOLI, BRUNO; BANISSI, E; KHOSROW-SHAHI, F; URSYN, A; BANNATYNE, MWM; PIRES, JM; et al. A class-based evaluation approach to assess multidimensional projections. 2020 24TH INTERNATIONAL CONFERENCE INFORMATION VISUALISATION (IV 2020), v. N/A, p. 8-pg., 2020-01-01. (18/25755-8, 18/17881-3)

MARCILIO-JR, WILSON ESTECIO; ELER, DANILO MEDEIROS. SADIRE: a context-preserving sampling technique for dimensionality reduction visualizations. JOURNAL OF VISUALIZATION, JUL 2020. (18/25755-8, 18/17881-3)

CHRISTOFANO, RAFAEL MARIANO; MARCILIO JUNIOR, WILSON ESTECIO; ELER, DANILO MEDEIROS; FILIPE, J; SMIALEK, M; BRODSKY, A; HAMMOUDI, S. PlaceProfile: Employing Visual and Cluster Analysis to Profile Regions based on Points of Interest. PROCEEDINGS OF THE 23RD INTERNATIONAL CONFERENCE ON ENTERPRISE INFORMATION SYSTEMS (ICEIS 2021), VOL 1, v. N/A, p. 9-pg., 2021-01-01. (18/25755-8, 18/17881-3)

MARCILIO-JR, WILSON E.; ELER, DANILO M.; GARCIA, ROGERIO E.. Contrastive analysis for scatterplot-based representations of dimensionality reduction. COMPUTERS & GRAPHICS-UK, v. 101, p. 46-58, DEC 2021. (18/25755-8, 18/17881-3)

MARCILIO-JR, WILSON ESTECIO; ELER, DANILO MEDEIROS. SADIRE: a context-preserving sampling technique for dimensionality reduction visualizations. JOURNAL OF VISUALIZATION, v. 23, n. 6, p. 999-1013, DEC 2020. (18/25755-8, 18/17881-3)

MARCILIO-JR, WILSON E.; ELER, DANILO; GUILHERME, IVAN; HURTER, C; PURCHASE, H; BOUATOUCH, K. Semi-automatic CNN Architectural Pruning using the Bayesian Case Model and Dimensionality Reduction Visualization. PROCEEDINGS OF THE 17TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS (VISAPP), VOL 4, v. N/A, p. 7-pg., 2022-01-01. (18/25755-8, 18/17881-3)