A5066927815- Data Visualization and Analytics
- Image Retrieval and Classification Techniques
- Computer Graphics and Visualization Techniques
- Video Analysis and Summarization
- Data Management and Algorithms
- Advanced Image and Video Retrieval Techniques
- Cell Image Analysis Techniques
- Advanced Text Analysis Techniques
- Music and Audio Processing
- Computational Geometry and Mesh Generation
- Tactile and Sensory Interactions
- Time Series Analysis and Forecasting
- Music Technology and Sound Studies
- Animal Vocal Communication and Behavior
- Complex Network Analysis Techniques
- Bioinformatics and Genomic Networks
- Natural Language Processing Techniques
- 3D Shape Modeling and Analysis
- Topic Modeling
- Interactive and Immersive Displays
- Digital Image Processing Techniques
- Advanced Clustering Algorithms Research
- Misinformation and Its Impacts
- Cellular Mechanics and Interactions
- Slime Mold and Myxomycetes Research
University College Cork
2020-2024
Makerere University
2024
Universidade de São Paulo
2011-2020
Brazilian Society of Computational and Applied Mathematics
2007-2020
Universidade Federal de São Carlos
2007-2016
Consejo Superior de Investigaciones Científicas
2001-2015
Hospital Universitário da Universidade de São Paulo
2014
University of East Anglia
1995-2002
United States University
2002
University of Lisbon
2000
Abstract Background Set comparisons permeate a large number of data analysis workflows, in particular workflows biological sciences. Venn diagrams are frequently employed for such but current tools limited. Results We have developed InteractiVenn, more flexible tool interacting with including up to six sets. It offers clean interface diagram construction and enables set unions while preserving the shape diagram. useful reveal differences similarities among sets may be guided our by tree or...
The problem of projecting multidimensional data into lower dimensions has been pursued by many researchers due to its potential application analysis various kinds. This paper presents a novel projection technique based on least square approximations. approximations compute the coordinates set projected points reduced number control with defined geometry. We name Least Square Projections (LSP). From an initial points, LSP defines positioning their neighboring through numerical solution that...
Abstract Different regions of oral squamous cell carcinoma (OSCC) have particular histopathological and molecular characteristics limiting the standard tumor−node−metastasis prognosis classification. Therefore, defining biological signatures that allow assessing prognostic outcomes for OSCC patients would be great clinical significance. Using histopathology-guided discovery proteomics, we analyze neoplastic islands stroma from invasive tumor front (ITF) inner to identify differentially...
Projection (or dimensionality reduction) techniques have been used as a means to handling the growing of data sets well providing way visualize information coded into point relationships. Their role is essential in interpretation and simultaneous use different projections their visualizations improve understanding increase level confidence result. For that purpose, should be fast allow multiple views same set. In this work we present novel technique for projecting multi-dimensional...
This paper presents a technique for generation of maps documents targeted at placing similar in the same neighborhood. As result, besides being able to group (and separate) by their contents, it runs very manageable computational costs. Based on multi-dimensional projection techniques and an algorithm improvement, results surface map that allows user identify number important relationships between sub-groups via visualization interaction. Visual attributes such as height, color, isolines...
Abstract Multidimensional projection has emerged as an important visualization tool in applications involving the visual analysis of high‐dimensional data. However, high precision methods are either computationally expensive or not flexible enough to enable feedback from user interaction into process. A built‐in mechanism that dynamically adapts based on direct intervention would make technique more useful for a larger range and data sets. In this paper we propose Piecewise Laplacian‐based...
Abstract Word clouds have become one of the most widely accepted visual resources for document analysis and visualization, motivating development several methods building layouts keywords extracted from textual data. Existing are effective to demonstrate content, but not capable preserving semantic relationships among while still linking word cloud underlying groups that generated them. Such representation is highly desirable exploratory collections. In this paper we present a novel approach...
Point placement strategies aim at mapping data points represented in higher dimensions to bi-dimensional spaces and are frequently used visualize relationships amongst instances.They have been valuable tools for analysis exploration of datasets various kinds. Many conventional techniques, however, do not behave well when the number is high, such as case documents collections. Later approaches handle that shortcoming, but may cause too much clutter allow flexible take place. In this work we...
Automatic data classification is a computationally intensive task that presents variable precision and considerably sensitive to the classifier configuration representation, particularly for evolving sets. Some of these issues can best be handled by methods support users' control over steps. In this paper, we propose visual methodology supports users in tasks related categorization such as training set selection; model creation, application verification; tuning. The approach then well suited...
Similarity-based layouts generated by multidimensional projections or other dimension reduction techniques are commonly used to visualize high-dimensional data. Many projection have been recently proposed addressing different objectives and application domains. Nonetheless, very little is known about the effectiveness of from a user's perspective, how distinct same data compare regarding typical visualization tasks they support, domain-specific issues affect outcome techniques. Learning more...
An alternative form to multidimensional projections for the visual analysis of data represented in spaces is deployment similarity trees, such as Neighbor Joining trees. They organize objects on plane emphasizing their levels with high capability detecting and separating groups subgroups objects. Besides this similarity-based hierarchical organization, some advantages include ability decrease point clutter; precision; a consistent view set during focusing, offering very intuitive way general...
This paper presents a tool, called text map explorer, which can be used to create and explore document maps (visual representations of collections). tool is capable grouping (and separating) documents by their contents, revealing the user relationships amongst them. also novel multi-dimensional projection technique for that reduces quadratic time complexity our previous approach O(N <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3/2</sup> ),...
Multidimensional projections map data points, defined in a high-dimensional space, into 1D, 2D or 3D representation space. Such mapping may be typically achieved with dimensional reduction, clustering, force directed point placement. Projections can displayed and navigated by analysts means of visual representations, which vary from points on plane to graphs, surfaces volumes. Typically, strive preserve distance relationships amongst as the original Information loss is inevitable projection...
The task of building effective representations to visualize and explore collections with moderate large number documents is hard. It depends on the evaluation some distance measure among texts also representation such relationships in bi- dimensional spaces. In this paper we introduce an alternative approach for visual maps based their content similarity, through reconstruction phylogenetic trees. tree capable representing that allows user quickly recover information detected by similarity...
Multidimensional projections (MPs) are key tools for the analysis of multidimensional data. MPs reduce data dimensionality while keeping original distance structure in low-dimensional output space, typically shown by a 2D scatterplot. While MP techniques grow more precise and scalable, they still do not show how dimensions (attributes) influence projection’s layout. In other words, which points similar, but why. We propose visual approach to describe contribute mostly similarity...
Abstract Visualization of high‐dimensional data requires a mapping to visual space. Whenever the goal is preserve similarity relations frequent strategy use 2D projections, which afford intuitive interactive exploration, e.g., by users locating and selecting groups gradually drilling down individual objects. In this paper, we propose framework for projecting 3D spaces, based on generalization Least‐Square Projection (LSP). We compare projections spaces both quantitatively through user study...
Document clustering is an efficient way to get insight into large text collections. Due the personalized nature of document clustering, even best fully automatic algorithms cannot create clusters that accurately reflect user»s perspectives. To incorporate perspective in process and, at same time, effectively visualize collections enhance user's sense-making data, we propose a novel visual analytics system for interactive clustering. We built our on top can adapt feedback. First, initial...