Feature selection aims to find the most important information from a given set of features. As this task can be seen as an optimization problem, combinatorial growth possible solutions may in-viable for exhaustive search. In paper we propose new nature-inspired feature technique based on bats behaviour, which has never been applied context so far. The wrapper approach combines power exploration together with speed Optimum-Path Forest classifier features that maximizes accuracy in validating...

Feature selection has been actively pursued in the last years, since to find most discriminative set of features can enhance recognition rates and also make feature extraction faster. In this paper, propose a new called Binary Cuckoo Search, which is based on behavior cuckoo birds. The experiments were carried out context theft detection power distribution systems two datasets obtained from Brazilian electrical company, have demonstrated robustness proposed technique against with several...

Parkinson's disease (PD) has affected millions of people world-wide, being its major problem the loss movements and, consequently, ability working and locomotion. Although we can find several works that attempt at dealing with this out there, most them make use datasets composed by a few subjects only. In work, present some results toward automated diagnosis PD means computer vision-based techniques in dataset dozens patients, which is one main contributions work. The part joint research...

Evolutionary algorithms have been widely used for Artificial Neural Networks (ANN) training, being the idea to update neurons' weights using social dynamics of living organisms in order decrease classification error. In this paper, we introduced Social-Spider Optimization improve training phase ANN with Multilayer perceptrons, and validated proposed approach context Parkinson's Disease recognition. The experimental section has carried out against five other well-known meta-heuristics...

Intradiffusion coefficients of 2,2,2-trifluoroethanol in water have been measured by the pulsed field gradient (PFG)-NMR spin–echo technique as a function temperature and composition on dilute alcohol region. The measurements extend range compositions already studied literature and, for first time, include study dependence. At same intradiffusion 2,2,2-trifluoroethanol, 2,2,3,3,3-pentafluoropropan-1-ol, 2,2,3,3,4,4,4-heptafluorobutan-1-ol were obtained computer simulation (molecular...

The world is facing a new era in which social media communication plays fundamental role people's lives. Along with irrefutable benefits, several collateral drawbacks have risen, one being the wide spread of false information malicious intents, what now commonly called "Fake News". fight against this problem not easy, especially when taking into account nature text messages involved on platforms (a sea small and myriad users). In work, we cope challenging authorship attribution posted...

According with the World Health Organization, around 50 million people in world have epilepsy. After diagnosis process, physicians classify epilepsy according to International Classification of Diseases, Ninth Revision (ICD-9). Often exams as electroencephalograms and magnetic resonances are used create a more accurate short amount time. The classification process is time consuming demands realization complementary exams. To circumvent this laborious we propose an automatic classifying...

The analysis of medical records is a major challenge, considering they are generally presented in plain text, have very specific technical vocabulary and nearly always unstructured. It an interdisciplinary work that requires knowledge from several fields. may goals, such as assistance on clinical decision, classification procedures, to support hospital management decisions. This presents the concepts involved, relevant existent related work, main open issues for future research within...

Meta-heuristic-based feature selection has been paramount in the last years, mainly because of its simplicity, effectiveness and also efficiency some cases. Such approaches are based on social dynamics living organisms, can vary from birds, bees, bats ants. Very recently, an optimization algorithm krill herd (KH) was proposed for continuous-valued applications, it more accurate than state-of-the-art techniques. In this paper, we propose a binary version KH technique, validate purposes...

The choice of hyper-parameters in Support Vector Machines (SVM)-based learning is a crucial task, since different values may degrade its performance, as well can increase the computational burden. In this paper, we introduce recently developed nature-inspired optimization algorithm to find out suitable for SVM kernel mapping named Social-Spider Optimization (SSO). We compare results obtained by SSO against with Grid-Search, Particle Swarm and Harmonic Search. Statistical evaluation has...

Clinical decision support systems play an important role in organizations. They have a tight relation with the information systems. Our goal is to develop system diagnosis and classification of epilepsy children. Around 50 million people world epilepsy. Epilepsy can be extremely complex process, demanding considerable time effort from physicians healthcare infrastructures. Exams such as electroencephalograms magnetic resonances are often used create more accurate short amount time. After...

Since the beginning, some pattern recognition techniques have faced problem of high computational burden for dataset learning. Among most widely used techniques, we may highlight Support Vector Machines (SVM), which obtained very promising results data classification. However, this classifier requires an expensive training phase, is dominated by a parameter optimization that aims to make SVM less prone errors over set. In paper, model finding such parameters as metaheuristic-based task,...

Epilepsy diagnosis can be an extremely complex process, demanding considerable time and effort from physicians healthcare infrastructures. Physicians need to classify each specific type of epilepsy based on different data, e.g., types seizures, events exams' results. This work presents a text mining approach support medical decisions relating classification in children. We propose process that, using patient records, applies ontologies named entities recognition as preprocessing steps, then...

Different events, such as terrorist acts and natural catastrophes, frequently occur across the world. The availability of images on internet can help to understand events. However, manually selecting representative (helpful) from a massive amount data be infeasible. Here, we propose an image semantic representation method that helps discrimination Representative Images (RI) Non-representative (NRI). Our method, called Event Semantic Space (ESS), generates low-dimensional by exploiting...