Hate speech detection in the Arabic language: corpus design, construction, and evaluation.

Hate Speech Detection in Arabic presents a multifaceted challenge due to the broad and diverse linguistic terrain. With its multiple dialects and rich cultural subtleties, Arabic requires particular measures to address hate speech online successfully. To address this issue, academics and developers have used natural language processing (NLP) methods and machine learning algorithms adapted to the complexities of Arabic text. However, many proposed methods were hampered by a lack of a comprehensive dataset/corpus of Arabic hate speech. In this research, we propose a novel multi-class public Arabic dataset comprised of 403,688 annotated tweets categorized as extremely positive, positive, neutral, or negative based on the presence of hate speech. Using our developed dataset, we additionally characterize the performance of multiple machine learning models for Hate speech identification in Arabic Jordanian dialect tweets. Specifically, the Word2Vec, TF-IDF, and AraBert text representation models have been applied to produce word vectors. With the help of these models, we can provide classification models with vectors representing text. After that, seven machine learning classifiers have been evaluated: Support Vector Machine (SVM), Logistic Regression (LR), Naive Bays (NB), Random Forest (RF), AdaBoost (Ada), XGBoost (XGB), and CatBoost (CatB). In light of this, the experimental evaluation revealed that, in this challenging and unstructured setting, our gathered and annotated datasets were rather efficient and generated encouraging assessment outcomes. This will enable academics to delve further into this crucial field of study.

Telemedicine interventions in type 2 diabetes management: a protocol for systematic review and network meta-analysis.

INTRODUCTION: The consequences of type 2 diabetes mellitus (T2DM) heavily strain individuals and healthcare systems worldwide. Interventions via telemedicine have become a potential tactic to tackle the difficulties in effectively managing T2DM. However, more research is needed to determine how telemedicine interventions affect T2DM management. This study sets out to systematically analyse and report the effects of telemedicine treatments on T2DM management to gain essential insights into the potential of telemedicine as a cutting-edge strategy to improve the outcomes and care delivery for people with T2DM. METHODS AND ANALYSIS: To uncover relevant research, we will perform a comprehensive literature search across six databases (PubMed, IEEE, EMBASE, Web of Science, Google Scholar and Cochrane Library). Each piece of data will be extracted separately, and any discrepancies will be worked out through discussion or by a third reviewer. The studies included are randomised controlled trial. We chose by predefined inclusion standards. After the telemedicine intervention, glycated haemoglobin will be the primary outcome. The Cochrane risk-of-bias approach will be used to evaluate the quality of the included studies. RevMan V.5.3.5 software and RStiduo V.4.3.1 software can be used to analyse the data, including publication bias. ETHICS AND DISSEMINATION: Since this research will employ publicly accessible documents, ethical approval is unnecessary. The review is registered prospectively on the PROSPERO database. The study's findings will be published in a peer-reviewed journal. PROSPERO REGISTRATION NUMBER: CRD42023421719.

PaCMAP-embedded convolutional neural network for multi-omics data integration.

Aims: The multi-omics data integration has emerged as a prominent avenue within the healthcare industry, presenting substantial potential for enhancing predictive models. The main motivation behind this study stems from the imperative need to advance prognostic methodologies in cancer diagnosis, an area where precision is pivotal for effective clinical decision-making. In this context, the present study introduces an innovative methodology that integrates copy number alteration (CNA), DNA methylation, and gene expression data. Methods: The three omics data were successfully merged into a two-dimensional (2D) map using the PaCMAP dimensionality reduction technique. Utilizing the RGB coloring scheme, a visual representation of the integration was produced utilizing the values of the three omics of each sample. Then, the colored 2D maps were fed into a convolutional neural network (CNN) to forecast the Gleason score. Results: Our proposed model outperforms the cutting-edge i-SOM-GSN model by integrating multi-omics data and the CNN architecture with an accuracy of 98.89, and AUC of 0.9996. Conclusion: This study demonstrates the effectiveness of multi-omics data integration in predicting health outcomes. The proposed methodology, combining PaCMAP for dimensionality reduction, RGB coloring for visualization, and CNN for prediction, offers a comprehensive framework for integrating heterogeneous omics data and improving predictive accuracy. These findings contribute to the advancement of personalized medicine and have the potential to aid in clinical decision-making for prostate cancer patients.

Software defect prediction using learning to rank approach.

Software defect prediction (SDP) plays a significant role in detecting the most likely defective software modules and optimizing the allocation of testing resources. In practice, though, project managers must not only identify defective modules, but also rank them in a specific order to optimize the resource allocation and minimize testing costs, especially for projects with limited budgets. This vital task can be accomplished using Learning to Rank (LTR) algorithm. This algorithm is a type of machine learning methodology that pursues two important tasks: prediction and learning. Although this algorithm is commonly used in information retrieval, it also presents high efficiency for other problems, like SDP. The LTR approach is mainly used in defect prediction to predict and rank the most likely buggy modules based on their bug count or bug density. This research paper conducts a comprehensive comparison study on the behavior of eight selected LTR models using two target variables: bug count and bug density. It also studies the effect of using imbalance learning and feature selection on the employed LTR models. The models are empirically evaluated using Fault Percentile Average. Our results show that using bug count as ranking criteria produces higher scores and more stable results across multiple experiment settings. Moreover, using imbalance learning has a positive impact for bug density, but on the other hand it leads to a negative impact for bug count. Lastly, using the feature selection does not show significant improvement for bug density, while there is no impact when bug count is used. Therefore, we conclude that using feature selection and imbalance learning with LTR does not come up with superior or significant results.

Comparing the diagnostic performance of ordinary, mixed, and lasso logistic regression models at identifying opioid and cannabinoid poisoning in U.S. dogs using pet demographic and clinical data reported to an animal poison control center (2005-2014).

Researchers have begun studying the impact of human opioid and cannabinoid use on dog populations. These studies have used data from an animal poison control center (APCC) and there are concerns that due to the illicit nature and social stigma concerning the use of these drugs, owners may not always be forthcoming with veterinarians or APCC staff regarding pet exposures to these toxicants. As a result, models derived from APCC data that examine the predictability of opioid and cannabinoid dog poisonings using pet demographic and health disorder information may help veterinarians or APCC staff more reliably identify these toxicants when examining or responding to a call concerning a dog poisoned by an unknown toxicant. The fitting of epidemiologically informed statistical models has been useful for identifying factors associated with various health conditions and as predictive tools. However, machine learning, including lasso regression, has many useful features as predictive tools, including the ability to incorporate large numbers of independent variables. Consequently, the objectives of our study were: 1) identify pet demographic and health disorders associated with opioid and cannabinoid dog poisonings using ordinary and mixed logistic regression models; and 2) compare the predictive performance of these models to analogous lasso logistic regression models. Data were obtained from reports of dog poisoning events collected by the American Society for the Prevention of Cruelty to Animals' (ASPCA) Animal Poisoning Control Center, from 2005-2014. We used ordinary and mixed logistic regression models as well as lasso logistic regression models with and without controlling for autocorrelation at the state level to train our models on half the dataset and test their predictive performance on the remainder. Although epidemiologically informed logistic regression models may require substantial knowledge of the disease systems being investigated, they had the same predictive abilities as lasso logistic regression models. All models had relatively high predictive parameters except for positive predictive values, due to the rare nature of calls concerning opioid and cannabinoid poisonings. Ordinary and mixed logistic regression models were also substantially more parsimonious than their lasso equivalents while still allowing for the epidemiological interpretation of model coefficients. Controlling for autocorrelation had little effect on the predictive performance of all models, but it did reduce the number of variables included in lasso models. Several disorder variables were associated with opioid and cannabinoid calls that were consistent with the acute effects of these toxicants. These models may help build diagnostic evidence concerning dog exposure to opioids and cannabinoids, saving time and resources when investigating these cases.

A Fecal-Microbial-Extracellular-Vesicles-Based Metabolomics Machine Learning Framework and Biomarker Discovery for Predicting Colorectal Cancer Patients.

Colorectal cancer (CRC) is one of the most common and lethal diseases among all types of cancer, and metabolites play a significant role in the development of this complex disease. This study aimed to identify potential biomarkers and targets in the diagnosis and treatment of CRC using high-throughput metabolomics. Metabolite data extracted from the feces of CRC patients and healthy volunteers were normalized with the median normalization and Pareto scale for multivariate analysis. Univariate ROC analysis, the t-test, and analysis of fold changes (FCs) were applied to identify biomarker candidate metabolites in CRC patients. Only metabolites that overlapped the two different statistical approaches (false-discovery-rate-corrected p-value < 0.05 and AUC > 0.70) were considered in the further analysis. Multivariate analysis was performed with biomarker candidate metabolites based on linear support vector machines (SVM), partial least squares discrimination analysis (PLS-DA), and random forests (RF). The model identified five biomarker candidate metabolites that were significantly and differently expressed (adjusted p-value < 0.05) in CRC patients compared to healthy controls. The metabolites were succinic acid, aminoisobutyric acid, butyric acid, isoleucine, and leucine. Aminoisobutyric acid was the metabolite with the highest discriminatory potential in CRC, with an AUC equal to 0.806 (95% CI = 0.700-0.897), and was down-regulated in CRC patients. The SVM model showed the most substantial discrimination capacity for the five metabolites selected in the CRC screening, with an AUC of 0.985 (95% CI: 0.94-1).

Explainable artificial intelligence model for identifying COVID-19 gene biomarkers.

AIM: COVID-19 has revealed the need for fast and reliable methods to assist clinicians in diagnosing the disease. This article presents a model that applies explainable artificial intelligence (XAI) methods based on machine learning techniques on COVID-19 metagenomic next-generation sequencing (mNGS) samples. METHODS: In the data set used in the study, there are 15,979 gene expressions of 234 patients with COVID-19 negative 141 (60.3%) and COVID-19 positive 93 (39.7%). The least absolute shrinkage and selection operator (LASSO) method was applied to select genes associated with COVID-19. Support Vector Machine - Synthetic Minority Oversampling Technique (SVM-SMOTE) method was used to handle the class imbalance problem. Logistics regression (LR), SVM, random forest (RF), and extreme gradient boosting (XGBoost) methods were constructed to predict COVID-19. An explainable approach based on local interpretable model-agnostic explanations (LIME) and SHAPley Additive exPlanations (SHAP) methods was applied to determine COVID-19- associated biomarker candidate genes and improve the final model's interpretability. RESULTS: For the diagnosis of COVID-19, the XGBoost (accuracy: 0.930) model outperformed the RF (accuracy: 0.912), SVM (accuracy: 0.877), and LR (accuracy: 0.912) models. As a result of the SHAP, the three most important genes associated with COVID-19 were IFI27, LGR6, and FAM83A. The results of LIME showed that especially the high level of IFI27 gene expression contributed to increasing the probability of positive class. CONCLUSIONS: The proposed model (XGBoost) was able to predict COVID-19 successfully. The results show that machine learning combined with LIME and SHAP can explain the biomarker prediction for COVID-19 and provide clinicians with an intuitive understanding and interpretability of the impact of risk factors in the model.

Phylogenomic Analyses and Molecular Signatures Elucidating the Evolutionary Relationships amongst the Chlorobia and Ignavibacteria Species: Robust Demarcation of Two Family-Level Clades within the Order Chlorobiales and Proposal for the Family Chloroherpetonaceae fam. nov.

Evolutionary relationships amongst Chlorobia and Ignavibacteria species/strains were examined using phylogenomic and comparative analyses of genome sequences. In a phylogenomic tree based on 282 conserved proteins, the named Chlorobia species formed a monophyletic clade containing two distinct subclades. One clade, encompassing the genera Chlorobaculum, Chlorobium, Pelodictyon, and Prosthecochloris, corresponds to the family Chlorobiaceae, whereas another clade, harboring Chloroherpeton thalassium, Candidatus Thermochlorobacter aerophilum, Candidatus Thermochlorobacteriaceae bacterium GBChlB, and Chlorobium sp. 445, is now proposed as a new family (Chloroherpetonaceae fam. nov). In parallel, our comparative genomic analyses have identified 47 conserved signature indels (CSIs) in diverse proteins that are exclusively present in members of the class Chlorobia or its two families, providing reliable means for identification. Two known Ignavibacteria species in our phylogenomic tree are found to group within a larger clade containing several Candidatus species and uncultured Chlorobi strains. A CSI in the SecY protein is uniquely shared by the species/strains from this "larger Ignavibacteria clade". Two additional CSIs, which are commonly shared by Chlorobia species and the "larger Ignavibacteria clade", support a specific relationship between these two groups. The newly identified molecular markers provide novel tools for genetic and biochemical studies and identification of these organisms.

Spatial, temporal, and space-time clusters associated with opioid and cannabis poisoning events in U.S. dogs (2005-2014).

While a substantial amount of research has focused on the abuse of opioids and cannabinoids in human populations, few studies have investigated accidental poisoning events in pet populations. The objective of this study was to identify whether poisoning events involving opioids and cannabinoids clustered in space, time, and space-time, and compare the locations of clusters between the two toxicants. Data were obtained concerning reports of dog poisoning events from the American Society for the Prevention of Cruelty to Animals' (ASPCA) Animal Poisoning Control Center (APCC), from 2005-2014. The spatial scan statistic was used to identify clusters with a high proportion of these poisoning events. Our analyses show that opioid and cannabinoid poisoning events clustered in space, time, and space-time. The cluster patterns identified for each toxicant were distinct, but both shared some similarities with human use data. This study may help increase awareness to the public, public health, and veterinary communities about where and when dogs were most affected by opioid and cannabinoid poisonings. This study highlights the need to educate dog owners about safeguarding opioid and cannabinoid products from vulnerable populations.

The impact of state cannabis legislation, county-level socioeconomic and dog-level characteristics on reported cannabis poisonings of companion dogs in the USA (2009-2014).

With current trends in cannabis legalization, large efforts are being made to understand the effects of less restricted legislation on human consumption, health, and abuse of these products. Little is known about the effects of cannabis legalization and increased cannabis use on vulnerable populations, such as dogs. The objective of this study was to examine the effects of different state-level cannabis legislation, county-level socioeconomic factors, and dog-level characteristics on dog cannabis poisoning reports to an animal poison control center (APCC). Data were obtained concerning reports of dog poisoning events, county characteristics, and state cannabis legislation from the American Society for the Prevention of Cruelty to Animals' (ASPCA) APCC, the US Census Bureau, and various public policy-oriented and government websites, respectively. A multilevel logistic regression model with random intercepts for county and state was fitted to investigate the associations between the odds of a call to the APCC being related to a dog being poisoned by a cannabis product and the following types of variables: dog characteristics, county-level socioeconomic characteristics, and the type of state-level cannabis legislation. There were significantly higher odds of a call being related to cannabis in states with lower penalties for cannabis use and possession. The odds of these calls were higher in counties with higher income variability, higher percentage of urban population, and among smaller, male, and intact dogs. These calls increased throughout the study period (2009-2014). Reporting of cannabis poisonings were more likely to come from veterinarians than dog owners. Reported dog poisonings due to cannabis appear to be influenced by dog-level and community-level factors. This study may increase awareness to the public, public health, and veterinary communities of the effects of recreational drug use on dog populations. This study highlights the need to educate dog owners about safeguarding cannabis products from vulnerable populations.

The identification of risk factors contributing to accidental opioid poisonings in companion dogs using data from a North American poison control center (2006-2014).

In the last decade, there has been a marked increase in opioid-related human deaths in the U.S. However, the effects of the growth in opioid use on vulnerable populations, such as pet dogs, are largely unknown. The objective of this study was to investigate potential risk factors at the dog, county, and state-levels that contributed to accidental dog opioid poisonings. Dog demographic information was collected during calls to the Animal Poison Control Center (APCC), operated by the American Society for the Prevention of Cruelty to Animals, about pet dog exposures to poisons from 2006-2014. Data concerning state-level opioid-related human death rates and county-level human opioid prescription rates were collected from databases accessed from the Centers for Disease Control and Prevention. A multilevel logistic regression model with random intercepts for county and state was fitted to explore associations between the odds of a call to the APCC being related to dog opioid poisonings with the following independent variables: sex, weight, age, reproductive status, breed class, year, source of calls, county-level human opioid prescription rate, and state-level opioid human death rate. There was a significant non-linear positive association between accidental opioid dog poisoning calls and county-level human opioid prescription rates. Similarly, the odds of a call being related to an opioid poisoning significantly declined over the study period. Depending on the breed class, the odds of a call being related to an opioid poisoning event were generally lower for older and heavier dogs. The odds of a call being related to an opioid poisoning were significantly higher for intact compared to neutered dogs, and if the call was made by a veterinarian compared to a member of the public. Veterinarians responding to poisonings may benefit from knowledge of trends in the use and abuse of both legal and illegal drugs in human populations.

Software Development Effort Estimation Using Regression Fuzzy Models.

Software effort estimation plays a critical role in project management. Erroneous results may lead to overestimating or underestimating effort, which can have catastrophic consequences on project resources. Machine-learning techniques are increasingly popular in the field. Fuzzy logic models, in particular, are widely used to deal with imprecise and inaccurate data. The main goal of this research was to design and compare three different fuzzy logic models for predicting software estimation effort: Mamdani, Sugeno with constant output, and Sugeno with linear output. To assist in the design of the fuzzy logic models, we conducted regression analysis, an approach we call "regression fuzzy logic." State-of-the-art and unbiased performance evaluation criteria such as standardized accuracy, effect size, and mean balanced relative error were used to evaluate the models, as well as statistical tests. Models were trained and tested using industrial projects from the International Software Benchmarking Standards Group (ISBSG) dataset. Results showed that data heteroscedasticity affected model performance. Fuzzy logic models were found to be very sensitive to outliers. We concluded that when regression analysis was used to design the model, the Sugeno fuzzy inference system with linear output outperformed the other models.

Phylogenetic analysis and molecular signatures defining a monophyletic clade of heterocystous cyanobacteria and identifying its closest relatives.

Detailed phylogenetic and comparative genomic analyses are reported on 140 genome sequenced cyanobacteria with the main focus on the heterocyst-differentiating cyanobacteria. In a phylogenetic tree for cyanobacteria based upon concatenated sequences for 32 conserved proteins, the available cyanobacteria formed 8-9 strongly supported clades at the highest level, which may correspond to the higher taxonomic clades of this phylum. One of these clades contained all heterocystous cyanobacteria; within this clade, the members exhibiting either true (Nostocales) or false (Stigonematales) branching of filaments were intermixed indicating that the division of the heterocysts-forming cyanobacteria into these two groups is not supported by phylogenetic considerations. However, in both the protein tree as well as in the 16S rRNA gene tree, the akinete-forming heterocystous cyanobacteria formed a distinct clade. Within this clade, the members which differentiate into hormogonia or those which lack this ability were also separated into distinct groups. A novel molecular signature identified in this work that is uniquely shared by the akinete-forming heterocystous cyanobacteria provides further evidence that the members of this group are specifically related and they shared a common ancestor exclusive of the other cyanobacteria. Detailed comparative analyses on protein sequences from the genomes of heterocystous cyanobacteria reported here have also identified eight conserved signature indels (CSIs) in proteins involved in a broad range of functions, and three conserved signature proteins, that are either uniquely or mainly found in all heterocysts-forming cyanobacteria, but generally not found in other cyanobacteria. These molecular markers provide novel means for the identification of heterocystous cyanobacteria, and they provide evidence of their monophyletic origin. Additionally, this work has also identified seven CSIs in other proteins which in addition to the heterocystous cyanobacteria are uniquely shared by two smaller clades of cyanobacteria, which form the successive outgroups of the clade comprising of the heterocystous cyanobacteria in the protein trees. Based upon their close relationship to the heterocystous cyanobacteria, the members of these clades are indicated to be the closest relatives of the heterocysts-forming cyanobacteria.