Multivariate and Online Prediction of Closing Price Using Kernel Adaptive Filtering

This paper proposes a multivariate and online prediction of stock prices via the paradigm of kernel adaptive filtering (KAF). The prediction of stock prices in traditional classification and regression problems needs independent and batch-oriented nature of training. In this article, we challenge this existing notion of the literature and propose an online kernel adaptive filtering-based approach to predict stock prices. We experiment with ten different KAF algorithms to analyze stocks’ performance and show the efficacy of the work presented here. In addition to this, and in contrast to the current literature, we look at granular level data. The experiments are performed with quotes gathered at the window of one minute, five minutes, ten minutes, fifteen minutes, twenty minutes, thirty minutes, one hour, and one day. These time windows represent some of the common windows frequently used by traders. The proposed framework is tested on 50 different stocks making up the Indian stock index: Nifty-50. The experimental results show that online learning and KAF is not only a good option, but practically speaking, they can be deployed in high-frequency trading as well.

Recognition of Electroencephalography-Related Features of Neuronal Network Organization in Patients With Schizophrenia Using the Generalized Choquet Integrals

In this study, we focused on the verification of suitable aggregation operators enabling accurate differentiation of selected neurophysiological features extracted from resting-state electroencephalographic recordings of patients who were diagnosed with schizophrenia (SZ) or healthy controls (HC). We built the Choquet integral-based operators using traditional classification results as an input to the procedure of establishing the fuzzy measure densities. The dataset applied in the study was a collection of variables characterizing the organization of the neural networks computed using the minimum spanning tree (MST) algorithms obtained from signal-spaced functional connectivity indicators and calculated separately for predefined frequency bands using classical linear Granger causality (GC) measure. In the series of numerical experiments, we reported the results of classification obtained using numerous generalizations of the Choquet integral and other aggregation functions, which were tested to find the most appropriate ones. The obtained results demonstrate that the classification accuracy can be increased by 1.81% using the extended versions of the Choquet integral called in the literature, namely, generalized Choquet integral or pre-aggregation operators.

Convergence of undulatory swimming kinematics across a diversity of fishes

Fishes exhibit an astounding diversity of locomotor behaviors from classic swimming with their body and fins to jumping, flying, walking, and burrowing. Fishes that use their body and caudal fin (BCF) during undulatory swimming have been traditionally divided into modes based on the length of the propulsive body wave and the ratio of head:tail oscillation amplitude: anguilliform, subcarangiform, carangiform, and thunniform. This classification was first proposed based on key morphological traits, such as body stiffness and elongation, to group fishes based on their expected swimming mechanics. Here, we present a comparative study of 44 diverse species quantifying the kinematics and morphology of BCF-swimming fishes. Our results reveal that most species we studied share similar oscillation amplitude during steady locomotion that can be modeled using a second-degree order polynomial. The length of the propulsive body wave was shorter for species classified as anguilliform and longer for those classified as thunniform, although substantial variability existed both within and among species. Moreover, there was no decrease in head:tail amplitude from the anguilliform to thunniform mode of locomotion as we expected from the traditional classification. While the expected swimming modes correlated with morphological traits, they did not accurately represent the kinematics of BCF locomotion. These results indicate that even fish species differing as substantially in morphology as tuna and eel exhibit statistically similar two-dimensional midline kinematics and point toward unifying locomotor hydrodynamic mechanisms that can serve as the basis for understanding aquatic locomotion and controlling biomimetic aquatic robots.

Low energy event classification in IceCube using boosted decision trees

The DeepCore sub-array within the IceCube Neutrino Observatory is a densely instrumented region of Antarctic ice designed to observe atmospheric neutrino interactions above 5 GeV via Cherenkov radiation. An essential aspect of any neutrino oscillation analysis is the ability to accurately identify the flavor of neutrino events in the detector. This task is particularly difficult at low energies when very little light is deposited in the detector. Here we discuss the use of machine learning to perform event classification at low energies in IceCube using a boosted decision tree (BDT). A BDT is trained using reconstructed quantities to identify track-like events, which result from muon neutrino charged current interactions. This new method improves the accuracy of particle identification compared to traditional classification methods which rely on univariate straight cuts.

Efficient Fruits Classification Using Convolutional Neural Network

Classification of fruits is a growing research topic in image processing. Various papers propose various techniques to deal with the classification of apples. However, some traditional classification methods remain drawbacks to producing an effective result with the big dataset. Inspired by deep learning in computer vision, we propose a novel learning method to construct a classification model, which can classify types of apples quickly and accurately. To conduct our experiment, we collect datasets, do preprocessing, train our model, tune parameter settings to get the highest accuracy results, then test the model using new data. Based on the experimental results, the classification model of green apples and red apples can obtain good accuracy with little loss. Therefore, the proposed model can be a promising solution to deal with apple classification.

Alpha Skew Gaussian Naïve Bayes Classifier

The main goal of this paper is to introduce a new procedure for a naïve Bayes classifier, namely alpha skew Gaussian naïve Bayes (ASGNB), which is based on a flexible generalization of the Gaussian distribution applied to continuous variables. As a direct advantage, this method can accommodate the possibility to handle with asymmetry in the uni or bimodal behavior. We provide the estimation procedure of this method, and we check the predictive performance when compared to other traditional classification methods using simulation studies and many real datasets with different application fields. The ASGNB is a powerful alternative to classification tasks when lie the presence of asymmetry of bimodality in the data and outperforms well when compared to other traditional classification methods in most of the cases analyzed.

Liar's war: Protecting civilians from disinformation during armed conflict

Disinformation in armed conflict may pose several distinctive forms of harm to civilians: exposure to retaliatory violence, distortion of information vital to securing human needs, and severe mental suffering. The gravity of these harms, along with the modern nature of wartime disinformation, is out of keeping with the traditional classification of disinformation in international humanitarian law (IHL) as a permissible ruse of war. A patchwork set of protections drawn from IHL, international human rights law and international criminal law may be used to limit disinformation operations during armed conflict, but numerous gaps and ambiguities undermine the force of this legal framework, calling for further scholarly attention and clarification.

SYSTEMATIC ANALYSIS OF THE SPECIES OF THE GENUS QUERCUS L. IN THE ARBORETUM HERBARIUM OF THE UNIVERSITY

The results of the analysis of Quercus L. species stored in the herbarium of the Department of Botany and Plant Physiology of Federal State Budget Educational Institution of Higher Education VSUFT (Voronezh) are presented. This herbarium of historical plant collections of the genus Quercus L. is critical for tracking changes in the genus, including the introduction and distribution of species. We examined the belonging of the species to systematic units – subgenus, section, subsection, row, using the traditional classification and the updated intrageneric classification of oaks. Information about the life form, plant height, date and place of collection of the specimen was recorded. We entered information into the database, which will further simplify the work on registration, revision of the herbarium fund and when replenishing herbarium specimens. Conclusions were drawn based on the results of the workabout changes in the taxonomic nature and phylogenetic relationships of species in Quercus L. genus