Using visual analytics for presenting comparative information on new drugs

2019

Multimed Tools Appl

Self Cite

Undirected graphs and symmetric square matrices are frequently found in various domains. An example is character co-occurrence matrices in digital humanities. However, the visualization of these datasets is difficult, especially if the graph is highly connected. In this article, we propose a method for visualizing undirected graphs and symmetric square matrices, by transforming them into overlapping sets, and then by visualizing these overlapping sets using set visualization techniques such as Euler diagram or rainbow boxes. We also propose a clustering approach to simplify the visualization.We apply this method to the visualization of various character co-occurrence matrices extracted from novels or DBpedia, ranging from 21 to 114 characters. We show that this visualization allows the finding of several interesting insights. Finally, we discuss the advantages and drawbacks of this method, and we compare it to other approaches in the literature.

Section: Rainbow Boxesmentioning

confidence: 99%

Visualizing undirected graphs and symmetric square matrices as overlapping sets

2019

Multimed Tools Appl

Self Cite

“…A more advanced program can take full advantage of the possibilities offered by Python for querying drug databases or generating graphical interfaces for data entering (before creating the contraindications) and for generating an HTML website displaying tables for comparing the contraindications (after the reasoning). In a subsequent work, we also used visual analytics for facilitating the comparison of drug properties [48].…”

Section: Use Case: Reasoning On Drug Contraindicationsmentioning

confidence: 99%

Owlready: Ontology-oriented programming in Python with automatic classification and high level constructs for biomedical ontologies

Artificial Intelligence in Medicine

2017

Self Cite

245

“…Rainbow boxes [14,13] are an information visualization technique that we recently proposed for overlapping sets. We applied this technique for the visualization of drug properties [15]. Several elements and sets are to be visualized, each element can belong to several sets and each set can contain several elements.…”

Section: Benchmarks and Testsmentioning

confidence: 99%

Artificial Feeding Birds (AFB): A New Metaheuristic Inspired by the Behavior of Pigeons

Advances in Nature-Inspired Computing and Applications

2018

Self Cite

Many optimization algorithms and metaheuristics have been inspired by nature. These algorithms often permit solving a wide range of optimization problems. Most of them were inspired by exceptional or extraordinary animal behaviors. On the contrary, in this chapter, we present Artificial Feeding Birds (AFB), a new metaheuristic inspired by the very trivial behavior of birds searching for food. AFB is very simple, yet efficient, and can be easily adapted to various optimization problems. We present application to unconstrained global nonlinear optimization, with several benchmark functions and the training of artificial neural networks (ANN), and to the resolution of ordering combinatorial optimization problems, with two examples: the traveling salesman problem and the optimization of rainbow boxes (a recent visualization technique for overlapping sets). We compare the results with those produced with Artificial Bee Colony (ABC), Firefly Algorithm (FA), Genetic Algorithm (GA) and Ant Colony Optimization (ACO), showing that AFB gives results equivalent or better than the other metaheuristics. Finally, we discuss the choice of inspiration sources from nature, before concluding.