Search Results (22)

This paper proposes a novel multiscale random forest model for stock index trend prediction, incorporating statistical inference principles to improve classification confidence. Traditional random forest classifiers rely on majority voting, which can yield biased estimates of class probabilities, especially under small sample sizes. To address this, we introduce a multiscale bootstrap correction mechanism into the ensemble framework, enabling the estimation of third-order accurate approximately unbiased p-values. This modification replaces naive voting with statistically grounded decision thresholds, improving the robustness of the model. Additionally, stepwise regression is employed for feature selection to enhance generalization. Experimental results on CSI 300 index data demonstrate that the proposed method consistently outperforms standard classifiers, including standard random forest, support vector machine, and weighted k-nearest neighbors model, across multiple performance metrics. The contribution of this work lies in the integration of hypothesis testing techniques into ensemble learning and the pioneering application of multiscale bootstrap inference to financial time series forecasting. Full article

Domain shift poses a major challenge in many Machine Learning applications due to variations in data acquisition protocols, particularly in the medical field. Test-time augmentation (TTA) can solve the domain shift issue and improve robustness by aggregating predictions from multiple augmented versions of the same input. However, TTA may inadvertently generate unrealistic or Out-of-Distribution (OOD) samples that negatively affect prediction quality. In this work, we introduce a filtering procedure that removes from the TTA images all the OOD samples whose representations lie far from the training data distribution. Moreover, all the retained TTA images are weighted inversely to their distance from the training data. The final prediction is provided by a Self-Ensemble with Confidence, which is a lightweight ensemble strategy that fuses predictions from the original and retained TTA samples using a weighted soft voting scheme, without requiring multiple models or retraining. This method is model-agnostic and can be integrated with any deep learning architecture, making it broadly applicable across various domains. Experiments on cross-domain leukocyte classification benchmarks demonstrate that our method consistently improves over standard TTA and Baseline inference, particularly when strong domain shifts are present. Ablation studies and statistical tests confirm the effectiveness and significance of each component. Full article

Background/Objectives: Musculoskeletal disorders causing chronic pain are increasingly prevalent due to factors such as injury, overuse, and aging, leading to interest in porcine collagen injections as a potential therapeutic and conservative option. Despite promising results, evidence-based information on this treatment is scarce. To address this gap, the authors conducted an eDelphi consensus among expert Italian physicians in musculoskeletal pain to gather their perspectives on collagen injections. Methods: A Steering Committee and a Panel of 23 physicians developed the statements list (36) including the modalities, safety, and efficacy of intra- and extra-articular collagen injections. Panelists rated their agreement with each statement on a 5-point Likert scale (5 means “Strong Agreement”). Consensus was defined as when at least 75% of the panelists voted with a score of ≥4/5 after two rounds of votes. The weighted average (WA) was calculated for each statement. As control, we elaborated a Hypothetical Parametric Distribution (HPD WA equal to 3.00), where the percent of panelists is equally distributed along each Likert Scale Value (LSV). The maximum WA for 75% of the consensus is established at 3.75. Indeed, the combination of 75% having WA > 3.75 was defined as “Strong Agreement”. While, if the consensus was under 75%, the WA vs. HPD comparison was performed using the Wilcoxon Test. Significant differences among the distribution of LSVs judged the statement as “Low Level of Agreement”. Disagreement was evaluated when the WA was under the PHD. Results: The consensus was reached “Strong Agreement” after twin rounds in 29 out of 36 (8.55%). In 5 out of 36 statements (13.89%), the panelists reached the “Low Level of Agreement” by statistical tests. In the remaining two statements, there was a “Consensus of Disagreement”. All panelists unanimously agreed on crucial points, such as contraindications, non-contraindication based solely on comorbidity, and the importance of monitoring collagen’s effectiveness. Unanimous agreement was reached on recommending ultrasound guidance and associating collagen injections with therapeutic exercise and physical modalities. Substantial consensus (concordance > 90%) supported collagen injections for osteoarthritis, chondropathy, and degenerative tendinopathies, emphasizing intra- and peri-articular treatment, even simultaneously. However, areas with limited evidence, such as the combination of collagen with other injectable drugs, treatment of myofascial syndrome, and injection frequency, showed disagreement. The potential of intra-tendinous porcine collagen injections for tendon regeneration yielded mixed results. Conclusions: Clinicians experts in musculoskeletal pain agree on using collagen injections to treat pain originating from joints (e.g., osteoarthritis) and periarticular (e.g., tendinopathies). Full article

Poor PQ is a partial case of power system impact on society and the environment. Although the significance of good PQ is generally understood, the topic has not yet been sufficiently explored in the scientific literature. Firstly, this paper discusses the role of PQ in sustainable development by distinguishing economic, environmental, and social parts, including the existing PQ impact assessment methods. PQ problems must be studied through such prisms as financial losses of industrial companies, damage to end-use equipment, natural phenomena, interaction with animals, and social issues related to law, people’s well-being, health and safety. Secondly, this paper presents the results of the survey of Lithuanian industrial companies, which focuses on the assessment of industrial equipment immunity to both voltage sags and supply interruptions, as well as a unique methodology based on expert assessment, IEEE Std 1564-2014 and EN 50160:2010 voltage sag tables, matrix theory, a statistical hypothesis test, and convolution-based sample comparison that was developed for this purpose. The survey was carried out during the PQ monitoring campaign in the Lithuanian DSO grid, and is one of the few PQ surveys presented in the scientific literature. After counting the votes and introducing the rating system (with and without weights), the samples are compared both qualitatively and quantitatively in order to determine whether the PQ impact on various end-use equipment is similar or not. Full article

Skin lesion classification plays a crucial role in dermatology, aiding in the early detection, diagnosis, and management of life-threatening malignant lesions. However, standalone transfer learning (TL) models failed to deliver optimal performance. In this study, we present an attention-enabled ensemble-based deep learning technique, a powerful, novel, and generalized method for extracting features for the classification of skin lesions. This technique holds significant promise in enhancing diagnostic accuracy by using seven pre-trained TL models for classification. Six ensemble-based DL (EBDL) models were created using stacking, softmax voting, and weighted average techniques. Furthermore, we investigated the attention mechanism as an effective paradigm and created seven attention-enabled transfer learning (aeTL) models before branching out to construct three attention-enabled ensemble-based DL (aeEBDL) models to create a reliable, adaptive, and generalized paradigm. The mean accuracy of the TL models is 95.30%, and the use of an ensemble-based paradigm increased it by 4.22%, to 99.52%. The aeTL models’ performance was superior to the TL models in accuracy by 3.01%, and aeEBDL models outperformed aeTL models by 1.29%. Statistical tests show significant p-value and Kappa coefficient along with a 99.6% reliability index for the aeEBDL models. The approach is highly effective and generalized for the classification of skin lesions. Full article

Malicious websites detection is one of the cyber-security tasks that protects sensitive information such as credit card details and login credentials from attackers. Machine learning (ML)-based methods have been commonly used in several applications of cyber-security research. Although there are some methods and approaches proposed in the state-of-the-art studies, the advancement of the most effective solution is still of research interest and needs to be improved. Recently, decision fusion methods play an important role in improving the accuracy of ML methods. They are broadly classified based on the type of fusion into a voting decision fusion technique and a divide and conquer decision fusion technique. In this paper, a decision fusion ensemble learning (DFEL) model is proposed based on voting technique for detecting malicious websites. It combines the predictions of three effective ensemble classifiers, namely, gradient boosting (GB) classifier, extreme gradient boosting (XGB) classifier, and random forest (RF) classifier. We use these classifiers because their advantages to perform well for class imbalanced and data with statistical noises such as in the case of malicious websites detection. A weighted majority-voting rule is utilized for generating the final decisions of used classifiers. The experimental results are conducted on a publicly available large dataset of malicious and benign websites. The comparative study exposed that the DFEL model achieves high accuracies, which are 97.25% on average of 10-fold cross-validation test and 98.50% on a holdout of 30% test set. This confirms the ability of proposed approach to improve the detection rate of malicious websites. Full article

Classical circular Hough transform was proven to be effective for some types of colorectal polyps. However, the polyps are very rarely perfectly circular, so some tolerance is needed, that can be ensured by applying fuzzy Hough transform instead of the classical one. In addition, the edge detection method, which is used as a preprocessing step of the Hough transforms, was changed from the generally used Canny method to Prewitt that detects fewer edge points outside of the polyp contours and also a smaller number of points to be transformed based on statistical data from three colonoscopy databases. According to the statistical study we performed, in the colonoscopy images the polyp contours usually belong to gradient domain of neither too large, nor too small gradients, though they can also have stronger or weaker segments. In order to prioritize the gradient domain typical for the polyps, a relative gradient-based thresholding as well as a gradient-weighted voting was introduced in this paper. For evaluating the improvement of the shape deviation tolerance of the classical and fuzzy Hough transforms, the maximum radial displacement and the average radius were used to characterize the roundness of the objects to be detected. The gradient thresholding proved to decrease the calculation time to less than 50% of the full Hough transforms, and the number of the resulting circles outside the polyp’s environment also decreased, especially for low resolution images. Full article

The root zone is located at the top level of the DNS system’s hierarchical structure and serves as the entry point for all domain name resolutions. The accuracy of the root zone file determines whether domain names can be resolved correctly. To solve the problems of single-source distrust and inaccurate data in the use of root zone files, this paper utilizes multi-source root zone files to build an accurate, real-time, and highly trustworthy root zone file through the validation of data accuracy and integrity. First, we propose a weighted voting statistical verification method. We select top-level domain name records with the highest confidence from the multi-source root zone data, thereby improving data accuracy. Second, through a dynamic cyclic construction process, we achieve dynamic monitoring of root zone file version changes, effectively ensuring the real-time nature of root zone data. Finally, we adopt a DNSSEC verification mechanism to address the issue of unreliable transmission paths for actively probed root zone data, ensuring data integrity by verifying the signed top-level domain name records and their ZSK, KSK keys. In addition, through the analysis of experimental data, we find that the main reason for the inaccuracy and unreliability of the root zone file is the delay in updating and synchronizing the file. We also discover the presence of redundant KSK keys in some of the source root zone data, which led to failure in the DNSSEC validation chain. The high-trust root zone file constructed in this paper provides data support for research on the root-side resolution anomaly detection and localization application of root zone files and has wide-ranging practical value. Full article

The films teenagers watch have a significant influence on their behavior. After witnessing a film starring an actor with a particular social habit or personality trait, viewers, particularly youngsters, may attempt to adopt the actor’s behavior. This study proposes an algorithm-based technique for predicting the market potential of upcoming science fiction films. Numerous science fiction films are released annually, and working in the film industry is both profitable and delightful. Before the film’s release, it is necessary to conduct research and make informed predictions about its success. In this investigation, different machine learning methods written in MATLAB are examined to identify and forecast the future performance of movies. Using 14 methods for machine learning, it was feasible to predict how individuals would vote on science fiction films. Due to their superior performance, the fine, medium, and weighted KNN algorithms were given more consideration. In comparison to earlier studies, the KNN-adopted methods displayed greater precision (0.89–0.93), recall (0.88–0.92), and accuracy (90.1–93.0%), as well as a rapid execution rate, more robust estimations, and a shorter execution time. These tabulated statistics illustrate that the weighted KNN method is effective and trustworthy. If several KNN algorithms targeting specific viewer behavior are logically coupled, the film business and its global expansion can benefit from precise and consistent forecast outcomes. This study illustrates how prospective data analytics could improve the film industry. It is possible to develop a model that predicts a film’s success, effect, and social behavior by assessing features that contribute to its success based on historical data. Full article

Two novel methods for heart disease prediction, which use the kurtosis of the features and the Maxwell–Boltzmann distribution, are presented. A Majority Voting approach is applied, and two base classifiers are derived through statistical weight calculation. First, exploitation of attribute kurtosis and attribute Kolmogorov–Smirnov test (KS test) result is done by plugging the base categorizer into a Bagging Classifier. Second, fitting Maxwell random variables to the components and summating KS statistics are used for weight assignment. We have compared state-of-the-art methods to the proposed classifiers and reported the results. According to the findings, our Gaussian distribution and kurtosis-based Majority Voting Bagging Classifier (GKMVB) and Maxwell Distribution-based Majority Voting Bagging Classifier (MKMVB) outperform SVM, ANN, and Naive Bayes algorithms. In this context, which also indicates, especially when we consider that the KS test and kurtosis hack is intuitive, that the proposed routine is promising. Following the state-of-the-art, the experiments were conducted on two well-known datasets of Heart Disease Prediction, namely Statlog, and Spectf. A comparison of Optimized Precision is made to prove the effectiveness of the methods: the newly proposed methods attained 85.6 and 81.0 for Statlog and Spectf, respectively (while the state of the heart attained 83.5 and 71.6, respectively). We claim that the Majority Voting family of classifiers is still open to new developments through appropriate weight assignment. This claim is obvious, especially when its simple structure is fused with the Ensemble Methods’ generalization ability and success. Full article

Social media platforms have become a substratum for people to enunciate their opinions and ideas across the globe. Due to anonymity preservation and freedom of expression, it is possible to humiliate individuals and groups, disregarding social etiquette online, inevitably proliferating and diversifying the incidents of cyberbullying and cyber hate speech. This intimidating problem has recently sought the attention of researchers and scholars worldwide. Still, the current practices to sift the online content and offset the hatred spread do not go far enough. One factor contributing to this is the recent prevalence of regional languages in social media, the dearth of language resources, and flexible detection approaches, specifically for low-resource languages. In this context, most existing studies are oriented towards traditional resource-rich languages and highlight a huge gap in recently embraced resource-poor languages. One such language currently adopted worldwide and more typically by South Asian users for textual communication on social networks is Roman Urdu. It is derived from Urdu and written using a Left-to-Right pattern and Roman scripting. This language elicits numerous computational challenges while performing natural language preprocessing tasks due to its inflections, derivations, lexical variations, and morphological richness. To alleviate this problem, this research proposes a cyberbullying detection approach for analyzing textual data in the Roman Urdu language based on advanced preprocessing methods, voting-based ensemble techniques, and machine learning algorithms. The study has extracted a vast number of features, including statistical features, word N-Grams, combined n-grams, and BOW model with TFIDF weighting in different experimental settings using GridSearchCV and cross-validation techniques. The detection approach has been designed to tackle users’ textual input by considering user-specific writing styles on social media in a colloquial and non-standard form. The experimental results show that SVM with embedded hybrid N-gram features produced the highest average accuracy of around 83%. Among the ensemble voting-based techniques, XGboost achieved the optimal accuracy of 79%. Both implicit and explicit Roman Urdu instances were evaluated, and the categorization of severity based on prediction probabilities was performed. Time complexity is also analyzed in terms of execution time, indicating that LR, using different parameters and feature combinations, is the fastest algorithm. The results are promising with respect to standard assessment metrics and indicate the feasibility of the proposed approach in cyberbullying detection for the Roman Urdu language. Full article

Blockchain technology has allowed many abnormal schemes to hide behind smart contracts. This causes serious financial losses, which adversely affects the blockchain. Machine learning technology has mainly been utilized to enable automatic detection of abnormal contract accounts in recent years. In spite of this, previous machine learning methods have suffered from a number of disadvantages: first, it is extremely difficult to identify features that enable accurate detection of abnormal contracts, and based on these features, statistical analysis is also ineffective. Second, they ignore the imbalances and repeatability of smart contract accounts, which often results in overfitting of the model. In this paper, we propose a data-driven robust method for detecting abnormal contract accounts over the Ethereum Blockchain. This method comprises hybrid features set by integrating opcode n-grams, transaction features, and term frequency-inverse document frequency source code features to train an ensemble classifier. The extra-trees and gradient boosting algorithms based on weighted soft voting are used to create an ensemble classifier that balances the weaknesses of individual classifiers in a given dataset. The abnormal and normal contract data are collected by analyzing the open source etherscan.io, and the problem of the imbalanced dataset is solved by performing the adaptive synthetic sampling. The empirical results demonstrate that the proposed individual feature sets are useful for detecting abnormal contract accounts. Meanwhile, combining all the features enhances the detection of abnormal contracts with significant accuracy. The experimental and comparative results show that the proposed method can distinguish abnormal contract accounts for the data-driven security of blockchain Ethereum with satisfactory performance metrics. Full article

Globally, coal remains one of the natural resources that provide power to the world. Thousands of people are involved in coal collection, processing, and transportation. Particulate coal dust is produced during these processes, which can crush the lung structure of workers and cause pneumoconiosis. There is no automated system for detecting and monitoring diseases in coal miners, except for specialist radiologists. This paper proposes ensemble learning techniques for detecting pneumoconiosis disease in chest X-ray radiographs (CXRs) using multiple deep learning models. Three ensemble learning techniques (simple averaging, multi-weighted averaging, and majority voting (MVOT)) were proposed to investigate performances using randomised cross-folds and leave-one-out cross-validations datasets. Five statistical measurements were used to compare the outcomes of the three investigations on the proposed integrated approach with state-of-the-art approaches from the literature for the same dataset. In the second investigation, the statistical combination was marginally enhanced in the ensemble of multi-weighted averaging on a robust model, CheXNet. However, in the third investigation, the same model elevated accuracies from 87.80 to 90.2%. The investigated results helped us identify a robust deep learning model and ensemble framework that outperformed others, achieving an accuracy of 91.50% in the automated detection of pneumoconiosis. Full article

Endometrial cancer (EC) is the most common gynecological neoplasm in high-income countries. Five-year survival rates are related to stage at diagnosis, but currently, no validated screening tests are available in clinical practice. The metabolome offers an unprecedented overview of the molecules underlying EC. In this study, we aimed to validate a metabolomics signature as a screening test for EC on a large study population of symptomatic women. Serum samples collected from women scheduled for gynecological surgery (n = 691) were separated into training (n = 90), test (n = 38), and validation (n = 563) sets. The training set was used to train seven classification models. The best classification performance during the training phase was the PLS-DA model (96% accuracy). The subsequent screening test was based on an ensemble machine learning algorithm that summed all the voting results of the seven classification models, statistically weighted by each models’ classification accuracy and confidence. The efficiency and accuracy of these models were evaluated using serum samples taken from 871 women who underwent endometrial biopsies. The EC serum metabolomes were characterized by lower levels of serine, glutamic acid, phenylalanine, and glyceraldehyde 3-phosphate. Our results illustrate that the serum metabolome can be an inexpensive, non-invasive, and accurate EC screening test. Full article

In recent years, there has been increasing concern about the effects of indoor thermal environments on human physical and mental health. This paper aimed to study the current status of the thermal environment and thermal comfort in the classrooms of Northeastern University during the heating season. The indoor thermal environment was analyzed with the use of field measurements, a subjective questionnaire, regression statistics, and the entropy weight method. The results show that personnel population density is an important factor affecting the temperature and relative humidity variations in classrooms. The results also show that the temperature and relative humidity in a lecture state are respectively 4.2 °C and 11.4% higher than those in an idle state. In addition, in university classrooms in Shenyang, the actual thermal neutral temperature is 2.5 °C lower than the predicted value of the Predicted Mean Vote. It was found that increasing indoor relative humidity can effectively improve the overall thermal comfort of subjects. Furthermore, the temperature preference of women was higher than that of men. Therefore, when setting the initial heating temperature, the personnel population density and sufficient indoor relative humidity have been identified as the key factors for improving the thermal environment of the classroom. Full article