Search

Purpose: In China, individuals with substance use disorders (SUD) face severe stigma, but reliable stigma assessment tool is lacking. Therefore, this study aimed to validate the Chinese version of the Substance Use Stigma Mechanism Scale (SU-SMS-C) and set its cut-off point. Methods: We recruited 1005 individuals with SUDs from Chinese rehabilitation centers. These participants completed a battery of questionnaires that included the SU-SMS-C, The Multidimensional Scale of Perceived Social Support (MSPSS), Center for Epidemiologic Studies Depression Scale (CES-D), General Self-Efficacy Scale (GSES), and Perceived Devaluation and Discrimination (PDD). Confirmatory factor analysis was used to assess the construct validity of the scale. Additionally, the Naive Bayes classifier was used to establish the cut-off point for the SU-SMS-C. We additionally explored the correlation between patient demographic characteristics and stigma. Results: A confirmatory factor analysis was utilized, revealing a second-order five-factor model. Based on the Naive Bayes classifier, the area under the receiver operating characteristic (AUCROC) of 0.746, the cut-off point for the SU-SMS-C was established at 44.5. The prevalence of stigma observed in the study population was 49.05%. Significant disparities were observed in the distribution of stigma across genders, with males experiencing more pronounced stigma than females. Moreover, patients consuming different primary substances reported diverse levels of stigma. Notably, those primarily using heroin endured a higher degree of stigma than users of other substances. Conclusion: The study is the first to identify a cut-off point for the SU-SMS-C by Naive Bayes classifier, bridging a major gap in stigma measurement research. SU-SMS-C may help treat and manage SUDs by reducing stigma. [ABSTRACT FROM AUTHOR]

Many applications of Natural Language Processing (NLP) like machine translation, document clustering, and information retrieval make use of Word Sense Disambiguation (WSD). WSD automatically predicts the sense of an ambiguous word that exactly fits it as per the given situation. While it may seem very easy for humans to interpret the meaning of natural language, machines require the processing of huge amounts of data for similar tasks. In this paper, we propose an automatic WSD system for the Kashmiri language based on the Naive Bayes classifier. This work is the first attempt towards developing a WSD system for the Kashmiri language to the best of our knowledge. Bag-of-Words (BoW) and Part-of-Speech (PoS) based features are used in this study for developing the WSD system. Experiments are carried out on a manually crafted sense-tagged dataset for 60 ambiguous Kashmiri words. These 60 words are selected based on the frequency in the raw corpus collected. Senses for annotation purposes of these ambiguous words are extracted from Kashmiri WordNet. The performance of the proposed system is measured using accuracy, precision, recall and F-1 measure metrics. The proposed WSD model reported the best performance (accuracy = 89.92, precision = 0.84, recall = 0.89, F-1 measure = 0.86) when both PoS and BoW features were used at the same time. [ABSTRACT FROM AUTHOR]

This paper explores the combination of a data mining-based state evaluation method for electrical equipment in substations, analyzing the effectiveness and accuracy. First, a Gaussian mixture model is applied to fit all raw data of electrical equipment. The Expectation Maximization algorithm summarizes the data distribution characteristics and identifies outliers. The a priori algorithm is then employed for data mining to derive frequent itemsets and association rules between equipment quality and measurement data. For new equipment samples, conditional probabilities of each feature are independently calculated and combined to classify and evaluate equipment quality. The results suggest that equipment reliability in smart substations can be inferred from historical and real-time operational data using improved association rule algorithms and Naive Bayes classifiers. Finally, the proposed method was applied to analyze statistical data from a 110 kV substation of a power supply company. The states prediction accuracy exceeded 95% when compared with actual equipment quality. The effectiveness evaluation metrics demonstrated that this method outperforms single-category algorithms in terms of accuracy and discrimination ability. [ABSTRACT FROM AUTHOR]

Aim: To explore biomarkers that can predict the response of type 2 diabetes (T2D) patients to metformin at an early stage to provide better treatment for T2D. Methods: T2D patients with (responders) or without response (non‐responders) to metformin were recruited, and their serum samples were used for metabolomic analysis to identify candidate biomarkers. Moreover, the efficacy of metformin was verified by insulin‐resistant mice, and the candidate biomarkers were verified to determine the biomarkers. Five different machine learning methods were used to construct the integrated biomarker profiling (IBP) with the biomarkers to predict the response of T2D patients to metformin. Results: A total of 73 responders and 63 non‐responders were recruited, and 88 differential metabolites were identified in the serum samples. After being verified in mice, 19 of the 88 were considered as candidate biomarkers. Next, after metformin regulation, nine candidate biomarkers were confirmed as the biomarkers. After comparing five machine learning models, the nine biomarkers were constructed into the IBP for predicting the response of T2D patients to metformin based on the Naïve Bayes classifier, which was verified with an accuracy of 89.70%. Conclusions: The IBP composed of nine biomarkers can be used to predict the response of T2D patients to metformin, enabling clinicians to start a combined medication strategy as soon as possible if T2D patients do not respond to metformin. [ABSTRACT FROM AUTHOR]

The 2010 Operational Concept for the Protection of Civilians (POC) has formalized the United Nations (UN) protection strategies into a three-tier structure: protection through dialogue and engagement (Tier 1), physical protection (Tier 2) and the establishment of a protective environment (Tier 3). Using a Naïve Bayes algorithm to classify 1,118 protection references in Security Council and General Assembly resolutions between 1990 and 2019, this article first systematically evaluates the impact of this policy. Has the three-tier approach shifted the traditional focus away from physical protection? Has the post-2010 period witnessed a surge in the prominence of holistic and long-term POC strategies? The findings reveal a notable increase in the frequency of protection mentions following the introduction of the Operational Concept. However, even in the final years of the observation period and in the context of five multi-dimensional peacekeeping operations, references to the physical protection cluster (Tier 2) outnumber mentions of Tiers 1 and 3. These findings hold policy relevance, as they stress the United Nations' difficulty in integrating the diverse protection approaches outlined in the three-tier structure of the Operational Concept into a cohesive strategy [ABSTRACT FROM AUTHOR]

The increase in fuel prices has significant impacts on the Indonesian community's economic sector. Most people object to this policy because of its significant effects on daily life. Additionally, Indonesia's economy has not fully recovered from the Covid-19 pandemic, compounded by news of rising oil prices. With news of the increase in fuel prices, most people express sentiment regarding the rise in fuel prices on one of the social media platforms, Twitter. This study aims to differentiate the sentiment provided by the public, whether positive, negative, or neutral, using the Naïve Bayes Classifier and Decision Tree algorithms. The analysis results show that the Naïve Bayes Algorithm model, specifically Bernoulli Naïve Bayes, achieves the highest accuracy of 65.60%, with a precision of 68%, recall of 60.30%, and f1-score of 59.33%. [ABSTRACT FROM AUTHOR]

One effort to reduce stunting rates is to predict stunting status early in toddlers. This study applies Naive Bayes (NB) to build a stunting prediction model because it is simple and easy to use. This study proposes a filter-based feature weighting technique to overcome the NB assumption, which states that each feature has the same contribution to the target. The frequency of an event in a dataset influences the feature weighting using mutual information criteria. This is the gap in the filter-based ranking highlighted in this study. Therefore, this study proposes a feature-weighting method that combines mutual information with the MOORA (MI-MOORA) decision-making method. This technique makes it possible to include external factors as criteria for ranking important features. For stunting cases, the external consideration for ranking purposes is the assessment of nutrition experts based on their experience in dealing with stunted toddlers. The MI-MOORA technique makes the availability of clean water the most influential feature that contributes to the stunting status. In the ten best features, the MI-MOORA ranking results are dominated by family factors. Based on the performance evaluation results of NB and other classifiers, MI-MOORA can improve the performance of stunt prediction models. [ABSTRACT FROM AUTHOR]

Introduction. The naïve Bayes classifier combined with principal component analysis allows to distinguish the effects of antipsychotic agents effectively as well as evaluate their dose-dependency based on their impact on electroencephalogram parameters in rats. Further development of this method requires its validation as an instrument for the screening of new understudied molecules. Valproic acid derivatives appear to be a promising neuropharmacological group as they exhibit not only antiepileptic activity but also mood-stabilizing, antimigraine, neuroprotective and analgesic effects.Aim. This work was carried out to perform the pharmacological screening of a valproic acid aminoester (AVA) that exhibits antidote properties in case of poisoning with anticholinesterase agents.Materials and methods. The experiments were conducted in white outbred rats with chronically implanted electrocorticographic electrodes. AVA was administered at doses of 0,5, 5 and 30 mg/kg. The training set, used as a reference to determine the pharmacological effects of each dose of the investigated substance, included matrixes of effects of 7 drugs: the antiepileptic agent sodium valproate, the D2-dopamine receptor blocker haloperidol, the M-cholinergic receptor blocker tropicamide, the H1-histamine receptor blocker chloropyramine, the acetylcholinesterase inhibitor galantamine, the sedative dexmedetomidine, and the anxiolytic phenazepam.Results and discussion. AVA at the dose of 0,5 mg/kg showed effects similar to those of sodium valproate, while a tenfold dose increase led to the predominance of an atropine-like effect. When administered at the dose of 30 mg/kg, the compound exhibited dexmedetomidine-like action. The central M-anticholinergic effect of AVA was confirmed by an arecoline test in mice, in which the substance at a dose of 88 mg/kg completely abolished the onset of tremor. Dexmedetomidine-like action was cancelled by the administration of atipamezole in equimolar quantities, which may indicate the potential сapability of AVA to activate α2-adrenergic receptors at a high dose. The results of molecular docking suggest that this effect is related specifically to the original aminoester molecule and not to its presumable active metabolites.Сonclusion. The obtained results confirm the effectiveness of the naïve Bayes classifier as an instrument for the prediction of the pharmacological activity of compounds based on their impact on electroencephalogram parameters in rats. Identification of new pharmacological effects of understudied compounds may widen the potential range of their clinical application as well as reveal probable adverse effects.

In recent years, machine learning (ML) has shown a significant impact in tackling various complicated problems in different application domains, including healthcare, economics, ecological, stock market, surveillance, and commercial applications. Machine Learning techniques are good enough to deal with a wide range of data, uncover fascinating links, offer insights, and spot trends. ML can improve disease diagnosis accuracy, predictability, performance, and reliability. This paper reviews various machine learning techniques applied to different medical datasets and proposes an ensemble method for helping in the early diagnosis of different diseases. The study compares existing machine learning techniques with the proposed ensemble method. The ensemble method uses the AdaBoost algorithm to combine the traits of choice trees, random forests, and support vector machines. Three feature selection techniques, Fisher’s score, information gain, and genetic algorithm, are used to select appropriate dataset features. The ensemble method also uses the K-fold cross-validation technique (where k=15) for validating results. SMOTE was employed to balance some of the datasets because they were quite unbalanced. All the methods used in this study are evaluated based on accuracy, AU Curve, Recall, Precision, and F1-score. The paper uses different medical datasets at the University of California Irvine and the Kaggle directory to compare machine-learning models with the proposed ensemble method. The encouraging results show that the ensemble method outperforms the existing machine-learning techniques. The paper thoroughly analyzes how machine learning is used in the medical industry, covering established technologies and their impact on medical diagnosis. An early diagnosis is needed to prevent people from deadly diseases. Hence, this study proposes an ensemble method that may be used to diagnose different diseases early. [ABSTRACT FROM AUTHOR]

The growing Internet of Things (IoT) landscape requires robust security; traditional rule-based systems are insufficient, driving the integration of machine learning (ML) for effective intrusion detection. This paper provides an inclusive overview of research efforts focused on harnessing ML methodologies to fortify intrusion detection within IoT. Tailored feature extraction techniques are pivotal for achieving high detection accuracy while minimizing false positives. The study employs the IoT23 dataset from Kaggle and incorporates four optimization algorithms – Particle Swarm Optimizer, Whale-Pearson optimization algorithm, Harris-Hawks Optimizer, and Support Vector Machine with Particle Swarm optimization algorithm (SVM-PSO) – for feature extraction and selection. A comparison with ML algorithms such as logistic regression, decision tree and naïve Bayes classifier highlights Harris-Hawks Optimizer as the most effective. Furthermore, ensemble methods, particularly the fusion of random forest with HHO optimization, yield an impressive accuracy of 99.97%, surpassing AdaBoost and XGBoost approaches. This paper underscores the application of diverse ensemble learning techniques to enhance intrusion detection precision and efficiency within the intricate IoT landscape, effectively tackling the challenges posed by its complex and ever-changing nature.

Taiwan's auditors have suffered from processing excessive audit data, including drawing audit evidence. This study advances sampling techniques by integrating machine learning with sampling. This machine learning integration helps avoid sampling bias, keep randomness and variability, and target risker samples. We first classify data using a Naive Bayes classifier into some classes. Next, a user-based, item-based, or hybrid approach is employed to draw audit evidence. The representativeness index is the primary metric for measuring its representativeness. The user-based approach samples data symmetrically around the median of a class as audit evidence. It may be equivalent to a combination of monetary and variable samplings. The item-based approach represents asymmetric sampling based on posterior probabilities for obtaining risky samples as audit evidence. It may be identical to a combination of non-statistical and monetary samplings. Auditors can hybridize those user-based and item-based approaches to balance representativeness and riskiness in selecting audit evidence. Three experiments show that sampling using machine learning integration has the benefits of drawing unbiased samples; handling complex patterns, correlations, and unstructured data; and improving efficiency in sampling big data. However, the limitations are the classification accuracy output by machine learning algorithms and the range of prior probabilities. [ABSTRACT FROM AUTHOR]

The growing Internet of Things (IoT) landscape requires robust security; traditional rule-based systems are insufficient, driving the integration of machine learning (ML) for effective intrusion detection. This paper provides an inclusive overview of research efforts focused on harnessing ML methodologies to fortify intrusion detection within IoT. Tailored feature extraction techniques are pivotal for achieving high detection accuracy while minimizing false positives. The study employs the IoT23 dataset from Kaggle and incorporates four optimization algorithms – Particle Swarm Optimizer, Whale-Pearson optimization algorithm, Harris-Hawks Optimizer, and Support Vector Machine with Particle Swarm optimization algorithm (SVM-PSO) – for feature extraction and selection. A comparison with ML algorithms such as logistic regression, decision tree and naïve Bayes classifier highlights Harris-Hawks Optimizer as the most effective. Furthermore, ensemble methods, particularly the fusion of random forest with HHO optimization, yield an impressive accuracy of 99.97%, surpassing AdaBoost and XGBoost approaches. This paper underscores the application of diverse ensemble learning techniques to enhance intrusion detection precision and efficiency within the intricate IoT landscape, effectively tackling the challenges posed by its complex and ever-changing nature. [ABSTRACT FROM AUTHOR]

This study explores machine learning's potential to classify carabao mangoes, a key Philippine export, into four grades based on size: A (large), B (medium), C (small), and R (reject). It introduces a Naïve Bayes classification model that uses image processing to extract features for grading. The goal is to create a consistent grading system to enhance export efficiency and benefit local farmers. The research aims to validate the Naïve Bayes model's accuracy using size, weight, area, and spot ratio. It employs a quantitative, experimental design, manipulating image processing techniques to gauge their impact on classification accuracy. The results show the Naïve Bayes model achieved 95% accuracy, effectively distinguishing large and reject mangoes. It performed well for medium and small mangoes, with a 7% error rate between these classes. This indicates the model's potential for quality control and sorting, though further refinement is needed to better differentiate between medium and small sizes. In conclusion, the study presents an image processing and Naïve Bayes-based method to classify carabao mangoes by size. The model's high accuracy suggests its effectiveness and potential for automating mango classification, which could significantly aid the Philippine mango industry. Further performance assessment was conducted using a confusion matrix. The research highlights the promise of this approach for efficient mango grading. [ABSTRACT FROM AUTHOR]

The rapid evolution of AI, especially in natural language processing, has given rise to conversational AI models like ChatGPT, revolutionizing user engagement through text interactions. This study focuses on sentiment analysis within ChatGPT interactions, uncovering emotional dynamics and evaluating sentiment analysis models. Sentiment analysis holds significance in diverse domains, aiding in user sentiment interpretation. The research employs systematic methods, including data collection, analysis in the PyCharm IDE, and hypothesis testing. Findings reveal a mixed sentiment dataset, with user prompts tending to be more positive than ChatGPT responses. Sentiment during conversations exhibits dynamic shifts, and a Naive Bayes classifier-based model shows robust performance. This research enhances our understanding of sentiment analysis in ChatGPT, offering insights for refining conversational AI and user experiences in an AI-driven world. [ABSTRACT FROM AUTHOR]

The rapid expansion of information resources on the internet has made it increasingly challenging for users to get valuable and relevant information. Recommendation systems (RSs) play a vital role in addressing this issue by providing personalized suggestions based on user preferences. However, the open nature of RSs allows for the injection of fake profiles by malicious users, leading to biased ratings and manipulation of the overall system results. These fake profiles have been intended to promote or degrade specific items, which creates bias in the system and compromises the user experience. The motivation behind these fake profiles is to promote or degrade a particular item, which affects the system and makes it more biased toward certain items. This research paper aims to detect and eliminate such fake profiles in RSs, ensuring users receive genuine and reliable results. The proposed approach utilizes a voting ensemble (VE) method, combining the strengths of multiple classification algorithms including Quadratic Discriminant Analysis (QDA), K-Nearest Neighbor (KNN), and Naive Bayes classifier (NBC). By utilizing these classifiers, the system enhances its robustness against malicious activities. The proposed VE method effectively identifies the fake profiles with a minimum number of parameters, and it shows high detection accuracy on Movielens datasets. It has been demonstrated that the proposed method gives 99.4%, 99.5%, 99.4% and 99.6% accuracy using NBC, QDA, KNN, and VE, respectively and it outperforms all other competing methods in terms of accuracy, precision, recall and MCC. [ABSTRACT FROM AUTHOR]

An increasing number of methods are being used to extract rock discontinuities from 3D point cloud data of rock surfaces. In this paper, a new method for automatic extraction of rock discontinuity based on an improved Naive Bayes classifier is proposed. The method first uses principal component analysis to find the normal vectors of the points, and then generates a certain number of random point sets around the selected training points for training the classifier. The trained, improved Naive Bayes classifier is based on point normal vectors and is able to automatically remove noise points due to various reasons in conjunction with the knee point algorithm, realizing high-precision extraction of the discontinuity sets. Subsequently, the individual discontinuities are segmented using a hierarchical density-based spatial clustering method with noise application. Finally, the PCA algorithm is used to complete the orientation by plane fitting the individual discontinuities. The method was applied in two cases, Kingston and Colorado, and the reliability and advantages of the new method were verified by comparing the results with those of previous research, and the discussion and analysis determined the optimal values of the relevant parameters in the algorithm. [ABSTRACT FROM AUTHOR]

Mobile Passport is an application that can be used as a digital service for people in Indonesia to apply for a new passport and an official online passport replacement from the Directorate General of Immigration replacing APAPO (Online Passport Service Application). User reviews of the Mobile Passport application are the output of big data generated as a result of the Internet of Things. The problem formulation in this research is how the implementation of the Naive Bayes text mining classifier algorithm can analyze the reviews contained in the Mobile Passport application as well as the accuracy, precision and recall values. This research uses the KDD (Knowledge Discovery and database) method which consists of data selection, data preprocessing, transformation, data mining, and evaluation using the R Studio tool. The resulting knowledge and information from this process is used as a useful knowledge base in decision making. The Naive Bayes classifier algorithm method in this research is used because of its reliability in handling data quickly and accurate predictions based on class probabilities, thus enabling research to obtain consistent and reliable results. [ABSTRACT FROM AUTHOR]

Visual observation to recognize and identify aerial objects is a means to protect air sovereignty, and the delay can endanger it. Visual observation can be done through Ground-to-Air (GTA) or by approaching through Air-to-Air (ATA) using binoculars. The observation needs accuracy and speed to speed up the decision-making process. We propose an Artificial Intelligence (AI) system that receives voice input to recognize and identify an Unmanned Aerial Aircraft (UAA). We employed Naïve Bayes Classifier (NBC) that processes inputs from voice-to-text tools containing the observed UAA's feature. With 70 UAA samples, the AI system achieved an accuracy of 79% and a WER of 4.16%. [ABSTRACT FROM AUTHOR]

In an era dominated by mobile communication, Short Message Service (SMS) plays a pivotal role in interpersonal interactions. However, the surge in unsolicited spam messages necessitates effective differentiation mechanisms. This exploratory data analysis (EDA) utilizes a dataset from the renowned UCI Machine Learning Repository to discern key characteristics distinguishing spam from legitimate messages. Employing Natural Language Processing (NLP) technique vectorization (BOW and TF-IDF), including the use of a Naïve Bayes algorithm and sentiment analysis, this investigation uncovers patterns and peculiarities specific to spam content. The findings highlight distinct differences in lexicon usage, message structure, and linguistic markers between spam and legitimate messages. For instance, spam messages often exhibit aggressive language and utilize unconventional structures. To elucidate, specific examples of such language patterns and structural anomalies are provided, offering a more nuanced understanding of the study's outcomes. Rooted in data-driven insights, this study lays the foundation for future endeavours in developing robust, NLP-powered spam detection mechanisms to preserve the essence of personal communication in the SMS sphere. Evaluating the model on a test dataset of 5,572 SMS messages yielded noteworthy results. The model demonstrated a precision rate of 98% for legitimate messages and an impeccable 100% precision for identifying spam without any false categorization. However, a notable dip in the recall rate for spam messages, at 85%, raises important considerations. This suggests potential challenges in detecting certain types of spam, emphasizing the need for further refinement in the model. The respective f1-scores for ham and spam messages were 99% and 92%, shedding light on the model's overall efficacy. These performance metrics not only quantify the model's accuracy at an admirable 98% but also prompt deeper reflections on the practical implications of the results, emphasizing areas for future research and enhancement in spam detection mechanisms within the dynamic landscape of mobile communication. [ABSTRACT FROM AUTHOR]

Although there has been significant progress in homomorphic encryption (HE) technology, a fully homomorphic Naive Bayes (NB) classifier capable of training on HE-encrypted data without decryption has not yet been efficiently developed. This research introduces a new method for approximating homomorphic logarithm calculations with an average relative error under 0.01%. Leveraging the SIMD functionality of the HE framework and a GPU, this technique can compute logarithm values for thousands of encrypted probabilities in about 2.5 seconds. Building upon this, we present a more efficient fully homomorphic NB classifier. Our method can train on a breast cancer dataset in roughly 14.3 seconds and perform query inferences in 0.84 seconds. Compared to the recent privacy-protecting NB classifier from Liu et al. in 2017, which offers a similar security level, our method is estimated to be about 28 times faster.

Chicken is one of the staple foods that is widely enjoyed by all. To obtain the benefits of chicken meat, the level of freshness becomes one of the main keys. In general, the level of freshness of chicken meat is divided into two classes, namely fresh and non-fresh. The difference in the level of freshness can be seen from the color changes of each class. Spoiled chicken (chicken died yesterday) is one type of meat in the non-fresh group. The widespread sale of spoiled chicken meat among the public raises doubts about choosing chicken that is suitable and unsuitable for consumption. Therefore, chicken meat freshness classification is needed to facilitate the selection of chicken meat based on color characteristics. The use of Naive Bayes Classifier algorithm in categorizing fresh and non-fresh classes is done by calculating the probability value of each image channel input. This research was conducted to compare the Naive Bayes, decision tree, and K-NN algorithms in classifying chicken meat based on color characteristics. The results of the study showed that the Naive Bayes classifier algorithm was superior to the decision tree and K-NN algorithms with an accuracy rate of 75%, precision of 79%, and recall of 65%. It is known that 27 images were predicted correctly and 9 images were predicted incorrectly out of a total 36 data. The use of a histogram in this study aims to differentiate chicken meat images from non-meat during the testing process of the model using the Naive Bayes classifier algorithm.

This paper explores the combination of a data mining-based state evaluation method for electrical equipment in substations, analyzing the effectiveness and accuracy. First, a Gaussian mixture model is applied to fit all raw data of electrical equipment. The Expectation Maximization algorithm summarizes the data distribution characteristics and identifies outliers. The a priori algorithm is then employed for data mining to derive frequent itemsets and association rules between equipment quality and measurement data. For new equipment samples, conditional probabilities of each feature are independently calculated and combined to classify and evaluate equipment quality. The results suggest that equipment reliability in smart substations can be inferred from historical and real-time operational data using improved association rule algorithms and Naive Bayes classifiers. Finally, the proposed method was applied to analyze statistical data from a 110 kV substation of a power supply company. The states prediction accuracy exceeded 95% when compared with actual equipment quality. The effectiveness evaluation metrics demonstrated that this method outperforms single-category algorithms in terms of accuracy and discrimination ability.

Introduction: A common task in the analysis of microbial communities involves assigning taxonomic labels to the sequences derived from organisms found in the communities. Frequently, such labels are assigned using machine learning algorithms that are trained to recognize individual taxonomic groups based on training data sets that comprise sequences with known taxonomic labels. Ideally, the training data should rely on labels that are experimentally verified--formal taxonomic labels require knowledge of physical and biochemical properties of organisms that cannot be directly inferred from sequence alone. However, the labels associated with sequences in biological databases are most commonly computational predictions which themselves may rely on computationallygenerated data--a process commonly referred to as "transitive annotation." Methods: In this manuscript we explore the implications of training a machine learning classifier (the Ribosomal Database Project's Bayesian classifier in our case) on data that itself has been computationally generated. We generate new training examples based on 16S rRNA data from a metagenomic experiment, and evaluate the extent to which the taxonomic labels predicted by the classifier change after re-training. Results: We demonstrate that even a few computationally-generated training data points can significantly skew the output of the classifier to the point where entire regions of the taxonomic space can be disturbed. Discussion and conclusions: We conclude with a discussion of key factors that affect the resilience of classifiers to transitively-annotated training data, and propose best practices to avoid the artifacts described in our paper. [ABSTRACT FROM AUTHOR]

This paper presents a study of the possibility of using machine learning methods to detect malicious installation files related to the type of Trojan installers and downloaders. A comparative analysis of machine learning algorithms applicable for the solution of this problem is provided: the naive Bayes classifier (NBC), random forest, and C4.5 algorithm. Machine learning models are developed using the Weka software. The most significant attributes of installation files of legitimate and Trojan programs are highlighted. [ABSTRACT FROM AUTHOR]

Cerebral microbleeds (CMBs) in the brain are the essential indicators of critical brain disorders such as dementia and ischemic stroke. Generally, CMBs are detected manually by experts, which is an exhaustive task with limited productivity. Since CMBs have complex morphological nature, manual detection is prone to errors. This paper presents a machine learning-based automated CMB detection technique in the brain susceptibility-weighted imaging (SWI) scans based on statistical feature extraction and classification. The proposed method consists of three steps: (1) removal of the skull and extraction of the brain; (2) thresholding for the extraction of initial candidates; and (3) extracting features and applying classification models such as random forest and naïve Bayes classifiers for the detection of true positive CMBs. The proposed technique is validated on a dataset consisting of 20 subjects. The dataset is divided into training data that consist of 14 subjects with 104 microbleeds and testing data that consist of 6 subjects with 63 microbleeds. We were able to achieve 85.7% sensitivity using the random forest classifier with 4.2 false positives per CMB, and the naïve Bayes classifier achieved 90.5% sensitivity with 5.5 false positives per CMB. The proposed technique outperformed many state-of-the-art methods proposed in previous studies. [ABSTRACT FROM AUTHOR]

Social media is one of the communication media that is widely used in the digital era as it is today. The use of social media allows people who are far apart to communicate and exchange media, both voice, video, and images quickly and even in real-time. In the past, the sale of protected animals was mostly done on the black market, usually involving a supply chain between sellers that usually existed in traditional markets or certain communities. With the existence of social media, the trend in conducting transactions and promoting wild animals has shifted from traditional to modern thanks to the support of existing technology. Protected wild animals are of concern to the local government or the global world to protect their existence. Therefore, this research proposes a machine learning (ML) based approach to detect the promotion and sale of wild animals on social media. The implementation of Naïve Bayes classifier (NBC) has a high accuracy in detecting trade in wild animals on social media with an accuracy value of 86. The implementation of ML-based approach is expected to produce new technology that allows authorities to know and monitor social media in order to reduce the sale and promotion of protected wildlife. [ABSTRACT FROM AUTHOR]

Facility location models involve identifying locations for facilities that provide services to the customers which are also called demand points. The p-median problem is a facility location problem which deals with locating p facilities in such a way that the sumtotal of demand-weighted distances between each demand point and its respective closest facility is minimized. The p-median problem does not take into account the possibility of failure of the facilities, but rather considers that the facilities once located will always be available to serve the customers or demand points. But, in reality, some of the facilities may face unpredicted disruptions, thereby forcing the customers to seek services from other functioning facilities. The reliability p-median problem (RpMP) concerns with locating facilities which minimize the cost while considering the cost of facility failures. In this paper, we have proposed a hyper-heuristic approach with naive Bayes classifier for the RpMP and compared its performance on standard benchmark instances with two best performing approaches for RpMP, viz. a genetic algorithm and a scatter search approach available in the literature. The results show the effectiveness of our approach in terms of the solution quality. [ABSTRACT FROM AUTHOR]

Pharmacoencephalography (pharmaco-EEG) is a prominent instrument for the pharmacological screening of new psychoactive molecules. This experimental approach has not remained a vestige of neurobiological studies, and can be used successfully to complete today's research objectives. The development and rise to universal use of machine learning techniques opens up novel prospects for the use of pharmaco-EEG data to solve the problems of classification and prognosis. We have previously shown that naïve Bayes classifier (NBC) combined with principal component analysis (PCA) can be used to differentiate between antipsychotic and sedative drug effects as well as to distinguish among the antipsychotics' effects. In the present study, we evaluated the possibility to employ this method to assess the dose-dependency of antipsychotic effects. The experiments were carried out in white outbred male rats with chronically implanted electrocorticographic electrodes. As the agents of interest, we chose two drugs with antipsychotic activity, chlorpromazine and promethazine, in three doses each (0.1, 1, 10 mg/kg and 0.5, 5 and 20 mg/kg, respectively). The training set, used as a reference to determine the pharmacological effects of the agents of interest, included the D2-dopamine receptor blocker haloperidol, M-cholinergic receptor blocker tropicamide, H1-histamine receptor blocker chloropyramine, the sedative dexmedetomidine, and the anxiolytic phenazepam. We have shown that the lowest chlorpromazine dose (0.1 mg/kg) can be characterized as antipsychotic with a marked histaminolytic effect, while the highest one (10 mg/kg) exhibits predominantly antipsychotic activity with a cataleptogenic effect. All three doses demonstrated anticholinergic activity, which increased with the dose. For promethazine, we observed a clear dose-dependent shift from antipsychotic action to cataleptogenic, alongside a notable antimuscarinic effect of all doses. None of promethazine doses showed any resemblance to chloropyramine, which probably indicates its anti-dopaminergic and antimuscarinic effects being able to mask its H1-antihistamine effect in the used dose range. In summary, our results demonstrate that NBC combined with PCA can be used to determine the dose-dependency of antipsychotic agents' effects based on their impact on electrocorticogram parameters. Further development of this method as well as expansion of psychotropic agent electropharmacogram library would allow for more precise prediction of pharmacological activity of the agents of interest. [ABSTRACT FROM AUTHOR]

Sentiment analysis is a particular form of content analysis, and its application has become popular with the growth of Internet platforms where a wide range of content is generated. Today, various classifiers use for sentiment analysis, and in this article, we show an example of using a Naïve Bayesian classifier. The aim is to examine the consistency of classifying textual materials into a positive, negative or neutral tone by analysts and the Bayesian algorithm. The hypotheses are that there is an increase in the agreement between the two ways of classifying textual materials as (1) the complexity of the formulations and (2) the size of the learning datasets increases. Based on the results, both hypotheses were accepted, but only on certain groups of messages. Increasing the size of the learning datasets and increasing the complexity of the formulations helped the classification accuracy for messages in a positive tone, while the classification accuracy for messages in other tones was high and equal regardless of varying the parameters. Correlation analysis showed a high positive correlation between the outcomes the Bayesian algorithm classified and the tones the analyst determined (r = 0,816). [ABSTRACT FROM AUTHOR]

Traffic accidents are still the main cause of fatalities, injuries and significant delays in highways. Understanding the accident contributing factor is imperative to increase safety in a traffic network. Recent research confirms that predictive modeling is an important tool to comprehend accident contributing factors. However, little effort has been put forward to explain complex machine learning models and their feature effects in accident prediction models. Thus, this study aims to build predictive models based on different machine learning methods and tries to explain the most contributing factors by using Shapley values which was developed based on game theory. Decision Trees, Neural Networks with Multilayer Perceptron (MLP), Support Vector Classifier, Case-Based Reasoning and Naive Bayes Classifier were used to predict the injury severity in accidents. Belt usage, alcohol consumption and speed violations were found as the most effective features and MLP gave the highest accuracy among all the applied predictive models. [ABSTRACT FROM AUTHOR]

The loading and unloading activities of a ship start from the activities of the ship berthing until the ship departs from the wharf. The arrival of ships to the wharf of a terminal has been scheduled or known as a window, but in reality there are still many delays in ship departure which result in an increase in the length of the ship berth so that it affects the low value of Box/Ship/Hour (BSH), and can result in the schedule of the next ship's schedule experiencing a delay berthing. This affects the income of a terminal and becomes the focus of attention for the terminal to be handled. This study was conducted to predict ship departure delay in the hope that it can help provide consideration to the terminal in compiling the next ship berthing schedule and the operational department can carry out anticipatory activities to prevent ship departure delay from ships that are predicted to be delayed through the allocation of loading and unloading facilities. The research was conducted by comparing several supervised learning methods, including the KNearest Neighbor (KNN), Naïve Bayes Classifier (NBC), NBC with Bagging Ensemble Classifier, Decision Tree, and Ordinal Logistic Regression. It was found that the best classification result is the KNN method (k = 5) with an accuracy value of 91%, but for the delay = 4 hours and > 4 hours it has a sensitivity value = 50%, it means that ships with a delay = 4 hours and > 4 hours have have less accurate predictions, and the use of the bagging method on the NBC method is proven to improve the classification performance of the NBC method. [ABSTRACT FROM AUTHOR]

Since the onset of the COVID-19 pandemic in Indonesia, many people have turned to telemedicine programs as an alternative to minimize social interactions, opting for consultations from the safety of their homes using smartphones and internet connectivity. Given the necessity for physical distancing and avoiding crowded places, these applications have become indispensable substitutes for in-person medical consultations. Numerous apps facilitating access to healthcare services have been introduced in Indonesia, ranging from business startups to initiatives by the Ministry of Health. Telemedicine can potentially revolutionize healthcare in Indonesia, addressing critical health challenges. A significant issue within Indonesia's healthcare system is the scarcity of doctors and their uneven distribution. With only four doctors per 10,000 people, this figure falls far below the WHO guideline of 10 doctors per 1,000. Sentiment analysis of these applications was conducted to evaluate how telemedicine applications meet public needs and offer an alternative solution. Lexicon-based and naive Bayes methods were employed to classify tweet data into positive, neutral, and negative sentiments. The results revealed 908 positive tweets, 172 negative tweets, and 168 neutral tweets, indicating predominantly positive public perceptions of telemedicine applications. The naive Bayes classifier exhibited a 74% accuracy rate, with a precision of 98% and a recall of 86%. These findings underscore the positive impact and acceptance of telemedicine applications among the Indonesian populace, emphasizing their significance in augmenting the nation's healthcare landscape.