Search

The {\em Kneser graph} $K(2n+k,n)$, for positive integers $n$ and $k$, is the graph $G=(V,E)$ such that $V=\{S\subseteq\{1,\ldots,2n+k\} : |S|=n\}$ and there is an edge $uv\in E$ whenever $u\cap v=\emptyset$. Kneser graphs have a nice combinatorial structure, and many parameters have been determined for them, such as the diameter, the chromatic number, the independence number, and, recently, the hull number (in the context of $P_3$-convexity). However, the determination of geodetic convexity parameters in Kneser graphs still remained open. In this work, we investigate both the geodetic number and the geodetic hull number of Kneser graphs. We give upper bounds and determine the exact value of these parameters for Kneser graphs of diameter two (which form a nontrivial subfamily). We prove that the geodetic hull number of a Kneser graph of diameter two is two, except for $K(5,2)$, $K(6,2)$, and $K(8,2)$, which have geodetic hull number three. We also contribute to the knowledge on Kneser graphs by presenting a characterization of endpoints of diametral paths in $K(2n+k,n)$, used as a tool for obtaining some of the main results in this work., Comment: 1 figure

Decision-support systems benefit from hidden patterns extracted from digital information. In the specific domain of gastropod characterization, morphometrical measurements support biologists in the identification of land snail specimens. Although snails can be easily identified by their excretory and reproductive systems, the after-death mollusk body is commonly inaccessible because of either soft material deterioration or fossilization. This study aims at characterizing Brazilian land snails by morphometrical data features manually taken from the shells. In particular, we examined a dataset of shells by using different learning models that labeled snail specimens with a precision up to 97.5% (F1-Score = .975, CKC = .967 and ROC Area = .998). The extracted patterns describe similarities and trends among land snail species and indicates possible outliers physiologies due to climate traits and breeding. Finally, we show some morphometrical characteristics dominate others according to different feature selection biases. Those data-based patterns can be applied to fast land snail identification whenever their bodies are unavailable, as in the recurrent cases of lost shells in nature or private and museum collections.

Background. The image-based identification of distinct tissues within dermatological wounds enhances patients' care since it requires no intrusive evaluations. This manuscript presents an approach, we named QTDU, that combines deep learning models with superpixel-driven segmentation methods for assessing the quality of tissues from dermatological ulcers. Method. QTDU consists of a three-stage pipeline for the obtaining of ulcer segmentation, tissues' labeling, and wounded area quantification. We set up our approach by using a real and annotated set of dermatological ulcers for training several deep learning models to the identification of ulcered superpixels. Results. Empirical evaluations on 179,572 superpixels divided into four classes showed QTDU accurately spot wounded tissues (AUC = 0.986, sensitivity = 0.97, and specificity = 0.974) and outperformed machine-learning approaches in up to 8.2% regarding F1-Score through fine-tuning of a ResNet-based model. Last, but not least, experimental evaluations also showed QTDU correctly quantified wounded tissue areas within a 0.089 Mean Absolute Error ratio. Conclusions. Results indicate QTDU effectiveness for both tissue segmentation and wounded area quantification tasks. When compared to existing machine-learning approaches, the combination of superpixels and deep learning models outperformed the competitors within strong significant levels.

Social media could provide valuable information to support decision making in crisis management, such as in accidents, explosions and fires. However, much of the data from social media are images, which are uploaded in a rate that makes it impossible for human beings to analyze them. Despite the many works on image analysis, there are no fire detection studies on social media. To fill this gap, we propose the use and evaluation of a broad set of content-based image retrieval and classification techniques for fire detection. Our main contributions are: (i) the development of the Fast-Fire Detection method (FFDnR), which combines feature extractor and evaluation functions to support instance-based learning, (ii) the construction of an annotated set of images with ground-truth depicting fire occurrences -- the FlickrFire dataset, and (iii) the evaluation of 36 efficient image descriptors for fire detection. Using real data from Flickr, our results showed that FFDnR was able to achieve a precision for fire detection comparable to that of human annotators. Therefore, our work shall provide a solid basis for further developments on monitoring images from social media., Comment: 12 pages, Proceedings of the International Conference on Enterprise Information Systems. Specifically: Marcos Bedo, Gustavo Blanco, Willian Oliveira, Mirela Cazzolato, Alceu Costa, Jose Rodrigues, Agma Traina, Caetano Traina, 2015, Techniques for effective and efficient fire detection from social media images, ICEIS, 34-45