Machine Learning Algorithms Decision Tree Research Articles

Precision phenotyping from routine laboratory parameters for out of hospital survival prediction in an all comers prospective PCI registry

Out-of-hospital mortality in coronary artery disease (CAD) is particularly high and established adverse event prediction tools are yet to be available. Our study aimed to investigate whether precision phenotyping can be performed using routine laboratory parameters for the prediction of out-of-hospital survival in a CAD population treated by percutaneous coronary intervention (PCI). All patients treated by PCI and discharged alive in a tertiary center between January 2016 – December 2022 that have been included prospectively in the local registry were analyzed. 115 parameters from the PCI registry and 266 parameters derived from routine laboratory testing were used. An extreme gradient-boosted decision tree machine learning (ML) algorithm was trained and used to predict all-cause and cardiovascular-cause survival. A total of 4027 patients with 4981 PCI hospitalizations were randomly included in the 70% training dataset and 1729 patients with 2160 PCI hospitalizations were randomly included in the 30% validation dataset. All-cause and cardiovascular cause mortality was 17.5% and 12.2%. The integrated area under the receiver operator characteristic curve for prediction of all-cause and cardiovascular cause mortality by the ML on the validation dataset was 0.844 and 0.837, respectively (all p < 0.001). Parameters reflecting renal function (first and maximum serum creatinine), hematologic function (mean corpuscular hemoglobin concentration, platelet distribution width), and inflammatory status (lymphocyte per monocyte ratio) were among the most important predictors. Accurate out-of-hospital survival prediction in CAD can be achieved using routine laboratory parameters. ML outperformed clinical risk scores in predicting out-of-hospital mortality in a prospective all-comers PCI population and has the potential to precisely phenotype patients.

ADVANCEMENTS IN AIRLINE SECURITY: EVALUATING MACHINE LEARNING MODELS FOR THREAT DETECTION

This study assessed the performance of four machine learning algorithms—Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and Neural Network (NN)—for predicting airline security threats using a dataset of 100,000 entries with 30 features. The models were evaluated based on accuracy, precision, recall, F1-Score, and AUC-ROC. The Neural Network achieved the highest performance, with an accuracy of 88%, precision of 86%, recall of 85%, F1-Score of 85.5%, and AUC-ROC of 0.90, demonstrating superior capability in capturing complex, non-linear patterns. The Random Forest model followed, with an accuracy of 85%, precision of 83%, recall of 82%, F1-Score of 82.5%, and AUC-ROC of 0.87, offering a robust and generalizable solution. The SVM model attained an accuracy of 81%, precision of 80%, recall of 78%, F1-Score of 79%, and AUC-ROC of 0.84, showing effective binary classification but with higher computational costs. The Decision Tree model, while interpretable, had the lowest performance metrics: accuracy of 78%, precision of 76%, recall of 72%, F1-Score of 74%, and AUC-ROC of 0.79. The results indicate that Neural Networks and Random Forests are the most effective models for airline security threat detection, with Neural Networks providing the highest overall accuracy and AUC-ROC.

Predicting immune risk in treatment-naïve HIV patients using a machine learning algorithm: a decision tree algorithm based on micronutrients and inversion of the CD4/CD8 ratio.

Micronutrients have significant functional implications for the human immune response, and the quality of food is a major factor affecting the severity and mortality caused by HIV in individuals undergoing antiretroviral therapy. A decrease in CD4 lymphocyte count and an increase in CD8 lymphocyte count are the hallmarks of HIV infection, which causes the CD4/CD8 ratio to invert from a normal value of >1.6 to <1.0. In this study, we tried to analyze whether the nutritional status of HIV-positive patients has an impact on the CD4/CD8 ratio inversion by utilizing a machine learning (ML) algorithm. In this study, 55 confirmed HIV-positive patients who had not started their anti-retroviral therapy were included after obtaining their informed, written consent. Moreover, 55 age-and sex-matched relatives and caregivers of the patients who tested negative in the screening were enrolled as controls. All individual patient data points were analyzed for model development with an 80-20 train-test split. Four trace elements, zinc (Zn), phosphate (P), magnesium (Mg), and calcium (Ca), were utilized by implementing a random forest classifier. The target of the study was the inverted CD4/CD8 ratio. The data of 110 participants were included in the analysis. The algorithm thus generated had a sensitivity of 80% and a specificity of 83%, with a likelihood ratio (LR+) of 4.8 and LR-of 0.24. The utilization of the ML algorithm adds to the limited evidence that currently exists regarding the role of micronutrients, especially trace elements, in the causation of immune risk. Our inherent strength lies in the fact that this study is one of the first studies to utilize an ML-based decision tree algorithm to classify immune risk in HIV patients. Our study uniquely corroborated the nutritional data to the immune risk in treatment-naïve HIV patients through the utilization of a decision tree ML algorithm. This could subsequently be an important classification and prognostic tool in the hands of clinicians.

A Predictive Model of Pressure Injury in Children Undergoing Living Donor Liver Transplantation Based on Machine Learning Algorithm.

The aim of our study was to formulate and validate a prediction model using machine learning algorithms to forecast the risk of pressure injuries (PIs) in children undergoing living donor liver transplantation (LDLT). A retrospective cohort study. The research was carried out at China's largest paediatric liver transplantation centre. A total of 438 children who underwent LDLT between June 2021 and December 2022 constituted the study cohort. The dataset was partitioned randomly into 70% for training datasets (306 cases) and 30% for testing datasets (132 cases). Utilising four machine learning algorithms-Decision Tree, Random Forest, Gradient Boosting Decision Tree and eXtreme Gradient Boosting-we identified risk factors and constructed predictive models. Out of 438 children, 42 developed PIs, yielding an incidence rate of 9.6%. Notably, 94% of these cases were categorised as Stage 1, and 54% were localised on the occiput. Upon evaluating the four prediction models, the Decision Tree model emerged as the most effective. The primary contributors to pressure injury in the Decision Tree model were identified as operation time, intraoperative corticosteroid administration, preoperative skin protection measures and preoperative skin conditions. A visualisation elucidating the logical inference process for the 10 variables within the Decision Tree model was presented. Ultimately, based on the Decision Tree model, a predictive system was developed. Machine learning algorithms facilitate the identification of crucial factors, enabling the creation of an effective Decision Tree model to forecast pressure injury development in children undergoing LDLT. With this predictive model at their disposal, nurses can assess the pressure injury risk level in children more intuitively. Subsequently, they can implement tailored preventive strategies to mitigate the occurrence of PIs. Paediatric patients contributed electronic health records datasets.

A holistic approach to software fault prediction with dynamic classification

Software Fault Prediction is a critical domain in machine learning aimed at pre-emptively identifying and mitigating software faults. This study addresses challenges related to imbalanced datasets and feature selection, significantly enhancing the effectiveness of fault prediction models. We mitigate class imbalance in the Unified Dataset using the Random-Over Sampling technique, resulting in superior accuracy for minority-class predictions. Additionally, we employ the innovative Ant-Colony Optimization algorithm (ACO) for feature selection, extracting pertinent features to amplify model performance. Recognizing the limitations of individual machine learning models, we introduce the Dynamic Classifier, a ground-breaking ensemble that combines predictions from multiple algorithms, elevating fault prediction precision. Model parameters are fine-tuned using the Grid-Search Method, achieving an accuracy of 94.129% and superior overall performance compared to random forest, decision tree and other standard machine learning algorithms. The core contribution of this study lies in the comparative analysis, pitting our Dynamic Classifier against Standard Algorithms using diverse performance metrics. The results unequivocally establish the Dynamic Classifier as a frontrunner, highlighting its prowess in fault prediction. In conclusion, this research introduces a comprehensive and innovative approach to software fault prediction. It pioneers the resolution of class imbalance, employs cutting-edge feature selection, and introduces dynamic ensemble classifiers. The proposed methodology, showcasing a significant advancement in performance over existing methods, illuminates the path toward developing more accurate and efficient fault prediction models.

Comparative Assessment of Fraudulent Financial Transactions using the Machine Learning Algorithms Decision Tree, Logistic Regression, Naïve Bayes, K-Nearest Neighbor, and Random Forest

Comparative Assessment of Fraudulent Financial Transactions using the Machine Learning Algorithms Decision Tree, Logistic Regression, Naïve Bayes, K-Nearest Neighbor, and Random Forest

P-201 Embryo prioritization through an AI-powered evaluation system improves the clinical outcome in IVF cycles

Abstract Study question How can the implantation rate be improved through the prioritization of embryos, particularly in the context of elective single embryo transfer (eSET)? Summary answer Based on the retrospective data of IVF treatment using 965 embryos, the AI-powered embryo-evaluation system, icONE, was developed to provide unbiased selection with promising accuracy. What is known already Over the past decade, numerous predictive models guiding IVF treatments for frozen-thawed embryo transfer (FET) success have emerged. However, most are unidimensional, focusing on factors like patient characteristics (e.g., age, reproductive hormones), embryo quality (e.g., morphokinetics), or biomarkers (e.g., genes, microRNAs). Despite efforts, many models exhibit limited AUROC (area under the receiver operating characteristic curve) values of 0.70∼0.77, suggesting a need for comprehensive consideration of maternal and embryonic factors to enhance predictive accuracy. Study design, size, duration The current study aims to develop an AI-assisted embryo prediction system. Nine hundred and sixty-five embryos from 232 patients (ages ranging from 24 to 50 years old, mean age = 36.9 years) who underwent IVF treatment during 2015–2018 were enrolled for developing an embryo prediction system. The clinical information and detail of IVF procedures of each patient and embryo were recorded and integrated to explore their impact on clinical outcomes. Participants/materials, setting, methods Pregnant outcomes were followed, including hCG examination, sac formation, and live birth. Whole genome sequencing (WGS) analysis using TE biopsy was performe. Genetic features extracted from WGS data and maternal clinicopathological characteristics were integrated and calculated by decision tree algorithm and machine learning algorithms, including Random Forest, XGBoost, LightGBM, Tabnet, etc. Main results and the role of chance To reduce the number of IVF cycles, diminish the risk associated with multiple pregnancy, and maximize the efficacy of IVF treatment, we analyzed genetic data generated from the preimplantation genetic screenings and developed a prototype of an AI-assisted program to evaluate the viability and quality of each embryo. The “intelligent cOmputing Noble Embryo” (icONE) program integrates genetic data and maternal clinical features. icONE showed classification accuracy and prediction specificity of 0.81 and 0.83, respectively, and the overall prediction accuracy of successful implantation reached to 92%. Random Forest outperformed other machine learning tools with overall positive predictive value, negative predictive value, specificity, and sensitivity of 91.8%, 91.0%, 90.3%, and 92.4%, respectively. The overall prediction accuracy of successful implantation reaches was 91.4%. Notably, the preliminary results of a randomized prospective clinical trial showed that the clinical pregnancy rate reached 77.3% (n = 81) based on the icONE’s evaluation, which was significantly higher than that of 50% (n = 85) in the non-icONE group. Patients of both advanced and young age benefited from icONE. Taken together, these results endorse the efficacy of icONE in prioritizing embryos for clinicians in IVF treatment. Limitations, reasons for caution It is noticed that our study is limited by the small sample size and all the data are collected from a single IVF center. The uniform and controlled operation of IVF cycles in a single center may cause selection bias. Wider implications of the findings Given the shift towards eSET in reproductive medicine, prioritizing embryos for implantation potential is vital in infertility treatment. The AI-powered model established in this study enhances clinical practice, optimizing parenthood chances for those facing fertility issues. Trial registration number not applicable

CLASSIFICATION OF KAZAKH MUSIC GENRES USING MACHINE LEARNING TECHNIQUES

This article analysis a Kazakh Music dataset, which consists of 800 audio tracks equally distributed across 5 different genres. The purpose of this research is to classify music genres by using machine learning algorithms Decision Tree Classifier and Logistic regression. Before the classification, the given data was pre-processed, missing or irrelevant data was removed. The given dataset was analyzed using a correlation matrix and data visualization to identify patterns. To reduce the dimension of the original dataset, the PCA method was used while maintaining variance. Several key studies aimed at analyzing and developing machine learning models applied to the classification of musical genres are reviewed. Cumulative explained variance was also plotted, which showed the maximum proportion (90%) of discrete values ​​generated from multiple individual samples taken along the Gaussian curve. A comparison of the decision tree model to a logistic regression showed that for f1 Score Logistic regression produced the best result for classical music - 82%, Decision tree classification - 75%. For other genres, the harmonic mean between precision and recall for the logistic regression model is equal to zero, which means that this model completely fails to classify the genres Zazz, Kazakh Rock, Kazakh hip hop, Kazakh pop music. Using the Decision tree classifier algorithm, the Zazz and Kazakh pop music genres were not recognized, but Kazakh Rock with an accuracy and completeness of 33%. Overall, the proposed model achieves an accuracy of 60% for the Decision Tree Classifier and 70% for the Logistic regression model on the training and validation sets. For uniform classification, the data were balanced and assessed using the cross-validation method. The approach used in this study may be useful in classifying different music genres based on audio data without relying on human listening.

High-performance grating-like SERS substrate based on machine learning for ultrasensitive detection of Zexie-Baizhu decoction

Rapid, universal and accurate identification of chemical composition changes in multi-component traditional Chinese medicine (TCM) decoction is a necessary condition for elucidating the effectiveness and mechanism of pharmacodynamic substances in TCM. In this paper, SERS technology, combined with grating-like SERS substrate and machine learning method, was used to establish an efficient and sensitive method for the detection of TCM decoction. Firstly, the grating-like substrate prepared by magnetron sputtering technology was served as a reliable SERS sensor for the identification of TCM decoction. The enhancement factor (EF) of 4-ATP probe molecules was as high as 1.90 × 107 and the limit of detection (LOD) was as low as 1 × 10−10 M. Then, SERS technology combined with support vector machine (SVM), decision tree (DT), Naive Bayes (NB) and other machine learning algorithms were used to classify and identify the three TCM decoctions, and the classification accuracy rate was as high as 97.78 %. In summary, it is expected that the proposed method combining SERS and machine learning method will have a high development in the practical application of multi-component analytes in TCM.

Analysis and Prediction of Students' Adaptation to Online Education Systems Based on Data Analysis and Decision Tree Machine Learning Algorithms

In today's digital age, the popularity and development of online education systems provide students with more flexible and convenient ways of learning. However, students' adaptation to the online education system is affected by a variety of factors, including gender, age, educational background, and field of specialisation. Through in-depth analyses and studies of these factors, the following conclusions can be drawn: gender has little influence on students' adaptation to online education, and male and female students perform similarly overall, but the proportion of male students at high adaptation levels is significantly higher than that of females. The majority of students show medium adaptability, indicating that the overall effect of online education is average. students in the age groups of 6-10, 16-20 and 26-30 years old have lower adaptability levels, and there are more low adaptability groups among students in colleges and universities. students majoring in IT are more adapted to the online education system, and students not majoring in IT have relatively poorer adaptability level. Local students are more adaptable to online education than foreign students. In areas with unstable electricity, students' adaptability is usually lower. The decision tree algorithm predictions showed good overall model accuracy, with higher prediction accuracy for students with high, low and medium levels of adaptability. The test set accuracy was 93.27%, and the precision and recall were both 93.33%, indicating excellent model predictions. In summary, by deeply analysing the influence of various factors on students' adaptation degree to online education and using the random forest algorithm to make predictions, it can provide an important reference for improving the effectiveness of online education systems and provide useful insights for personalised education.

Multiple Disease Prediction Using Machine Learning

Machine learning, which is a type of computer technology, has changed healthcare a lot. It helps doctors predict diseases better and faster. In healthcare, using machine learning algorithms decision tree (DT), logistic regression (LR), support vector machine (SVM) that can help predict lots of different diseases at the same time. This helps doctors find and treat illnesses early, which makes patients better and saves money on healthcare. This paper looks at how we can use computer programs that learn from data to predict many diseases. It talks about why this is good, what problems we might face, and where we might go next with it. We give a summary of the several machine learning models and information sources that are often employed in illness prediction. We also go over the significance of feature selection, model assessment, and combining several data modalities for improved illness prediction. We give a summary of the several machine learning models and information sources that are often employed in illness prediction. We also go over the significance of feature selection, model assessment, and combining several data modalities for improved illness prediction. The research shows that using machine learning algorithms to predict many diseases at once could really help public health. Again, we use a machine learning model to determine whether or not an individual is impacted by a few diseases. This training model trains itself to predict illness using sample data.

Creditworthiness pattern prediction and detection for GCC Islamic banks using machine learning techniques

PurposeBank creditworthiness refers to the evaluation of a bank’s ability to meet its financial obligations. It is an assessment of the bank’s financial health, stability and capacity to manage risks. This paper aims to investigate the credit rating patterns that are crucial for assessing creditworthiness of the Islamic banks, thereby evaluating the stability of their industry.Design/methodology/approachThree distinct machine learning algorithms are exploited and evaluated for the desired objective. This research initially uses the decision tree machine learning algorithm as a base learner conducting an in-depth comparison with the ensemble decision tree and Random Forest. Subsequently, the Apriori algorithm is deployed to uncover the most significant attributes impacting a bank’s credit rating. To appraise the previously elucidated models, a ten-fold cross-validation method is applied. This method involves segmenting the data sets into ten folds, with nine used for training and one for testing alternatively ten times changeable. This approach aims to mitigate any potential biases that could arise during the learning and training phases. Following this process, the accuracy is assessed and depicted in a confusion matrix as outlined in the methodology section.FindingsThe findings of this investigation reveal that the Random Forest machine learning algorithm superperforms others, achieving an impressive 90.5% accuracy in predicting credit ratings. Notably, our research sheds light on the significance of the loan-to-deposit ratio as a primary attribute affecting credit rating predictions. Moreover, this study uncovers additional pivotal banking features that intensely impact the measurements under study. This paper’s findings provide evidence that the loan-to-deposit ratio looks to be the purest bank attribute that affects credit rating prediction. In addition, deposit-to-assets ratio and profit sharing investment account ratio criteria are found to be effective in credit rating prediction and the ownership structure criterion came to be viewed as one of the essential bank attributes in credit rating prediction.Originality/valueThese findings contribute significant evidence to the understanding of attributes that strongly influence credit rating predictions within the banking sector. This study uniquely contributes by uncovering patterns that have not been previously documented in the literature, broadening our understanding in this field.

Investigation of light gas adsorption by microporous materials using data analysis and decision tree machine learning algorithm

This study aimed to extract insights into the adsorption behavior of light gases by microporous materials such as zeolites, MOFs, and activated carbons. Data reported in 22 articles published between 1974 and 2022 were analyzed using a decision tree (DT) machine learning algorithm. A comprehensive database comprising 3297 data points, elucidating the impacts of 8 input variables on adsorption capacity, was constructed. Various exploratory data analysis techniques, including histograms, bar charts, and scatter plots, were employed to discern the relationships among input variables and the performance variable. Additionally, a DT model was employed to regress the adsorbed capacity data. Furthermore, a parametric study of this model facilitated the determination of the relative importance of input variables and their partial dependence, enhancing the interpretability of the model and enabling the deduction of heuristics for high or low adsorption capacity. The exploratory data analysis revealed that pressure and molecular mass of the adsorbate were the most significant variables influencing adsorption capacity.

Predicting and analyzing the fill factor of non-fullerene organic solar cells based on material properties and interpretable machine-learning strategies

In recent years, encouraging progress in non-fullerene acceptors-based organic solar cells (NFAs-OSCs) has been made in the field of clean-energy technologies. Nevertheless, achieving a high fill factor (FF) for NFAs-OSCs is a great challenge due to the FF value can be dramatically affected by various factors. An accurate prediction (based on empirical data) of how inherent characteristics of polymer-NFA combinations (e.g., frontier molecular orbitals (FMO) and charge-carrier mobilities) affects the FF in NFAs-OSCs is therefore urgently needed for the future commercialization prospect. The presented work demonstrates an outperforming predictive model using the optimized Gradient-boosting decision tree (GBDT) machine learning (ML) algorithm to accurately predict the FF value, which is based on a dataset consisting of > 180 unique donor/NFA blends with reported FF from previously published publications and experimental physical descriptors (FMO and charge-carrier mobilities). In addition, the SHapley Additive exPlanations (SHAP) strategy is further applied to not only gain the interpretability of the GBDT model prediction but also directly map the relationship between the FF value and experimental physical descriptors by systematic interpretable explanations (e.g., global and local). The SHAP analysis qualitatively reveals that the highest occupied molecular orbital (HOMO) of the donor and the HOMO of the acceptor are the two most important features associated with FF value prediction, which is in good agreement with previous experimental facts. Furthermore, the ML strategies explore the key requirements (e.g., small HOMO offset (<0.3 eV), high hole and electron mobilities (>1.0 × 10–4 cm2 V−1 s−1), and desired μh/μe ratio (in the range of 1.8–3.3)) for enhancing the FF values of NFAs-OSCs. This study provides novel insights into the possible physical mechanisms of FF enhancement, thus facilitating research efforts (e.g., rational design of energy level tuning in donor/NFA blend) for high-performance NFAs-OSCs.

Analyzing electric vehicle battery health performance using supervised machine learning

Lithium-ion batteries having high energy and power densities, fast depleting cost, and multifaceted technological improvement lead to the first choice for electric transportation systems. A noble supervised K nearest neighbors (KNN), support vector regressor (SVR), decision tree (DT), and random forest (RF) regressors machine learning algorithm are developed with different accuracy and usefulness for the state of health estimation from direct measurable indices of voltage, current, and temperature without inherited electrochemical characteristics like voltage hysteresis, aging, degradation level, operational, and environmental effect. Based on the battery dataset from Sandia National Laboratories, the proposed technique is validated. The comparison characteristics are provided with the help of mean absolute percent error (MAPE), mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE) techniques. The MAPE error at 15 °C for different regression algorithms SVR, DT, KNN, and RF are 2.791E-02, 1.607E-03, 9.957E-04, and 4.323E-03 respectively. The MAE error at 25 °C for different regression algorithms SVR, DT, KNN, and RF are 5.593E-02, 2.379E-03, 2.429E-03, and 5.073E-03 respectively. The RMSE error at 35 °C for different regression algorithms SVR, DT, KNN, and RF are 5.062E-02, 6.862E-03, 6.333E-03, and 1.275E-02 respectively. The MSE error percentages for the testing sample at 35 °C for different regression algorithms are 0.37 %, 0.01 %, 0.01 %, and 0.02 % respectively. The results indicate that KNN and DT exhibit higher precision when applied to lithium-ion batteries for electric vehicles. Collaborating with hardware manufacturers to exchange data has the potential to foster the creation of innovative battery health monitoring systems.

Mobile Customer Satisfaction Scoring Research Based on Quadratic Dimension Reduction and Machine Learning Integration

Customer satisfaction is a measure of the degree of satisfaction of customer experience. Among the three major operators in China, China Mobile plays an important role in the communication field. A study of customer satisfaction with China Mobile will have a significant positive impact on the sustainable development of the entire communication industry. In order to respond to customer needs accurately, a mobile customer satisfaction research method based on quadratic dimensionality reduction and machine learning integration is proposed. Firstly, the core evaluation system of impact satisfaction is established, through the integration of systematic clustering and exploratory factor analysis for quadratic dimensionality reduction. Then, unreasonable data in the core influencing factors are eliminated. Finally, the gradient-boosted decision tree (GBDT) machine learning algorithm is applied to predict satisfaction, with a prediction accuracy of up to 99%, and the highly accurate satisfaction prediction can quickly respond to customer needs and feedback to improve customer experience and satisfaction.

Habitat in flames: How climate change will affect fire risk across koala forests

Aim:Generate fire susceptibility maps for the present and 2070, to identify the threat wildfires pose to koalas now and under future climate change. Location:Australia. Time period:Present and 2070. Major taxa studied:60 main tree species browsed by koalas. Method:The Decision Tree machine learning algorithm was applied to generate a fire susceptibility index (a measure of the potential for a given area or region to experience wildfires) using a dataset of conditioning factors, namely: altitude, aspect, rainfall, distance from rivers, distance from roads, forest type, geology, koala presence and future dietary sources, land use-land cover (LULC), normalized difference vegetation index (NDVI), slope, soil, temperature, and wind speed. Results:We found a general increase in susceptibility of Australian vegetation to bushfires overall. The simulation for current conditions indicated that 39.56% of total koala habitat has a fire susceptibility rating of “very high” or “high”, increasing to 44.61% by 2070. Main conclusions:Wildfires will increasingly impact koala populations in the future. If this iconic and vulnerable marsupial is to be protected, conservation strategies need to be adapted to deal with this threat. It is crucial to strike a balance between ensuring that koala habitats and populations are not completely destroyed by fire while also allowing for forest rejuvenation and regeneration through periodic burns.

A Pseudorange-Based GPS Spoofing Detection Using Hyperbola Equations

Spoofing attacks are one of the most critical threats against secure global navigation satellite system (GNSS) positioning. Since correct positioning is a must in many vehicle-to-everything (V2X) communication systems, the detection of these attacks is vital. In the literature on the detection strategies of spoofing attacks, the majority of the solutions are based on the assumption that all available signals are spoofed. In this paper, we focus on detecting spoofing attacks in which authentic and spoofing positioning signals coexist in a V2X system. The observable pseudorange values are utilized with the help of derived hyperbola equations during the design of the spoofing detection algorithms. We propose an algorithm, which is named as the sub-optimal search-based spoofing detection algorithm (Algorithm 1), and it considers all possible numbers of spoofing attacking signals, but not all spoofing scenarios with the same amount of spoofing signals. To address the complexity problems based on the increased number of search scenarios of this approach, we propose another algorithm, which is called subset selection-based spoofing detection algorithm (Algorithm 2), with a smart selection of the search subsets. Both of these algorithms are first compared with fixed detection thresholds, which are determined with the Pareto front approach. Then, the performance of the algorithms is investigated when vehicle mobility and spoofing imperfection are considered. Finally, a supervised learning-based decision tree machine learning (ML) algorithm is run without specifying any detection threshold. The results indicate that Algorithm 1 provides higher detection rates than the subset selection-based algorithm; however, the false alarm ratios of Algorithm 2 are much lower than its original performance.

Decision Tree Models and Machine Learning Algorithms in the Fault Recognition on Power Lines with Branches

The complication of the structure, topology and composition of the future electrical networks is characterized by difficult-to-recognize circuit-mode situations and requires modern methods for analyzing information parameters. The growing trend of digitizing signals in substations and the use of the IEC 61850 standard results in a huge amount of new data available at the nodes of the electrical network. The development and analysis of new methods for detecting and recognizing the modes of electrical networks (normal and emergency) are topical research issues. The article explores a new approach to recognizing a faulted section of an electrical network with branches by concurrently analyzing several information features and applying machine learning methods: decision tree, random forest, and gradient boosting. The application of this approach for decision-making by relay protection has not been previously implemented. Simulation modeling and the Monte Carlo method are at the heart of obtaining training samples. The results of testing the studied methods under review showed the required flexibility, the ability to use a large number of information parameters, as well as the best results of fault recognition in comparison with the distance protection relay.

Geometric graph learning with extended atom-types features for protein-ligand binding affinity prediction

Understanding and accurately predicting protein-ligand binding affinity are essential in the drug design and discovery process. At present, machine learning-based methodologies are gaining popularity as a means of predicting binding affinity due to their efficiency and accuracy, as well as the increasing availability of structural and binding affinity data for protein-ligand complexes. In biomolecular studies, graph theory has been widely applied since graphs can be used to model molecules or molecular complexes in a natural manner. In the present work, we upgrade the graph-based learners for the study of protein-ligand interactions by integrating extensive atom types such as SYBYL and extended connectivity interactive features (ECIF) into multiscale weighted colored graphs (MWCG). By pairing with the gradient boosting decision tree (GBDT) machine learning algorithm, our approach results in two different methods, namely sybylGGL-Score and ecifGGL-Score. Both of our models are extensively validated in their scoring power using three commonly used benchmark datasets in the drug design area, namely CASF-2007, CASF-2013, and CASF-2016. The performance of our best model sybylGGL-Score is compared with other state-of-the-art models in the binding affinity prediction for each benchmark. While both of our models achieve state-of-the-art results, the SYBYL atom-type model sybylGGL-Score outperforms other methods by a wide margin in all benchmarks. Finally, the best-performing SYBYL atom-type model is evaluated on two test sets that are independent of CASF benchmarks.