Communication between people from different communities can sometimes be hampered by the lack of knowledge of each other's language. A large number of people needs to learn a language in order to ensure a fluid communication or want to do it just out of intellectual curiosity. To assist language learners' needs tutor tools have been developed. In this paper we present a tutor for learning the basic 42 hand configurations of the Spanish Sign Language, as well as more than one hundred of common words. This tutor registers the user image from an off-the-shelf webcam and challenges her to perform the hand configuration she chooses to practice. The system looks for the configuration, out of the 42 in its database, closest to the configuration performed by the user, and shows it to her, to help her to improve through knowledge of her errors in real time. The similarities between configurations are computed using Procrustes analysis. A table with the most frequent mistakes is also recorded and available to the user. The user may advance to choose a word and practice the hand configurations needed for that word. Sign languages have been historically neglected and deaf people still face important challenges in their daily activities. This research is a first step in the development of a Spanish Sign Language tutor and the tool is available as open source. A multidimensional scaling analysis of the clustering of the 42 hand configurations induced by Procrustes similarity is also presented.

Abstract Natural gestures are a desirable feature for a humanoid robot, as they are presumed to elicit a more comfortable interaction in people. With this aim in mind, we present in this paper a system to develop a natural talking gesture generation behavior. A Generative Adversarial Network (GAN) produces novel beat gestures from the data captured from recordings of human talking. The data is obtained without the need for any kind of wearable, as a motion capture system properly estimates the position of the limbs/joints involved in human expressive talking behavior. After testing in a Pepper robot, it is shown that the system is able to generate natural gestures during large talking periods without becoming repetitive. This approach is computationally more demanding than previous work, therefore a comparison is made in order to evaluate the improvements. This comparison is made by calculating some common measures about the end effectors’ trajectories (jerk and path lengths) and complemented by the Fréchet Gesture Distance (FGD) that aims to measure the fidelity of the generated gestures with respect to the provided ones. Results show that the described system is able to learn natural gestures just by observation and improves the one developed with a simpler motion capture system. The quantitative results are sustained by questionnaire based human evaluation .

BACKGROUND: Microarray technology provides the expression level of many genes. Nowadays, an important issue is to select a small number of informative differentially expressed genes that provide biological knowledge and may be key elements for a disease. With the increasing volume of data generated by modern biomedical studies, software is required for effective identification of differentially expressed genes. Here, we describe an R package, called ORdensity, that implements a recent methodology (Irigoien and Arenas, 2018) developed in order to identify differentially expressed genes. The benefits of parallel implementation are discussed. RESULTS: ORdensity gives the user the list of genes identified as differentially expressed genes in an easy and comprehensible way. The experimentation carried out in an off-the-self computer with the parallel execution enabled shows an improvement in run-time. This implementation may also lead to an important use of memory load. Results previously obtained with simulated and real data indicated that the procedure implemented in the package is robust and suitable for differentially expressed genes identification. CONCLUSIONS: The new package, ORdensity, offers a friendly and easy way to identify differentially expressed genes, which is very useful for users not familiar with programming. AVAILABILITY: https://github.com/rsait/ORdensity.

This paper presents a novel approach for iris dissimilarity computation based on Machine Learning paradigms and Computer Vision transformations. Based on the training dataset given by the MICHE II Challenge organizers, a set of classifiers has been constructed and tested, aiming at classifying a single image. The main novelty of this paper remains in the used approach to iris dissimilarity computation: given two iris images, both of them are classified using the same paradigm, obtaining the a posteriori probability for each of the considered class values. Hence, two distributions are obtained, one for each iris image, and the dissimilarity is computed as the distance between these two distributions. Experimental results indicate the appropriateness of this new approach, even though more research and experiments are needed to obtain some improvements and to accelerate the classification process.

This paper summarizes the proposal submitted by the joint team conformed by researchers from UPV and ULPGC to the Mobile Iris CHallenge Evaluation II. The approach makes use of a state-of-the-art iris segmentation technique, to later extract features making use of local descriptors. Those suitable to the problem are selected after evaluating a collection of 15 local descriptors, covering a range of different grid configuration setups. A Machine Learning approach is used, learning a supervised classifier to deal with the descriptors data. A classifier is obtained for each descriptor, and the best ones are combined in a multi-classifier system. The final step fuses the classifier outputs obtained for 5 different local descriptors, to compute the dissimilarity measure for a pair of iris images.

Online communities are becoming an important tool in the communication and participation processes in our society. However, the most widespread applications are difficult to use for people with disabilities, or may involve some risks if no previous training has been undertaken. This work describes a novel social network for cognitively disabled people along with a clustering‐based method for modeling activity and socialization processes of its users in a noninvasive way. This closed social network is specifically designed for people with cognitive disabilities, called Guremintza, that provides the network administrators (e.g., social workers) with two types of reports: summary statistics of the network usage and behavior patterns discovered by a data mining process. Experiments made in an initial stage of the network show that the discovered patterns are meaningful to the social workers and they find them useful in monitoring the progress of the users.

Class binarization strategies decompose the original multi-class problem into several binary sub-problems. One versus One (OVO) is one of the most popular class binarization techniques, which considers every pair of classes as a different sub-problem. Usually, the same classifier is applied to every sub-problem and then all the outputs are combined by some voting scheme. In this paper we present a novel idea where for each test instance we try to assign the best classifier in each sub-problem of OVO. To do so, we have used two simple Dynamic Classifier Selection (DCS) strategies that have not been yet used in this context. The two DCS strategies use K-NN to obtain the local region of the test-instance, and the classifier that performs the best for those instances in the local region, is selected to classify the new test instance. The difference between the two DCS strategies remains in the weight of the instance. In this paper we have also proposed a novel approach in those DCS strategies. We propose to use the K-Nearest Neighbor Equality (K-NNE) method to obtain the local accuracy. K-NNE is an extension of K-NN in which all the classes are treated independently: the K nearest neighbors belonging to each class are selected. In this way all the classes take part in the final decision. We have carried out an empirical study over several UCI databases, which shows the robustness of our proposal.