Extreme Gradient Boosting 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.

Machine Learning Based Prediction of Imminent ICP Insults During Neurocritical Care of Traumatic Brain Injury.

In neurointensive care, increased intracranial pressure (ICP) is a feared secondary brain insult in traumatic brain injury (TBI). A system that predicts ICP insults before they emerge may facilitate early optimization of the physiology, which may in turn lead to stopping the predicted ICP insult from occurring. The aim of this study was to evaluate the performance of different artificial intelligence models in predicting the risk of ICP insults. The models were trained to predict risk of ICP insults starting within 30min, using the Uppsala high frequency TBI dataset. A restricted dataset consisting of only monitoring data were used, and an unrestricted dataset using monitoring data as well as clinical data, demographic data, and radiological evaluations was used. Four different model classes were compared: Gaussian process regression, logistic regression, random forest classifier, and Extreme Gradient Boosted decision trees (XGBoost). Six hundred and two patients with TBI were included (total monitoring 138,411h). On the task of predicting upcoming ICP insults, the Gaussian process regression model performed similarly on the Uppsala high frequency TBI dataset (sensitivity 93.2%, specificity 93.9%, area under the receiver operating characteristic curve [AUROC] 98.3%), as in earlier smaller studies. Using a more flexible model (XGBoost) resulted in a comparable performance (sensitivity 93.8%, specificity 94.6%, AUROC 98.7%). Adding more clinical variables and features further improved the performance of the models slightly (XGBoost: sensitivity 94.1%, specificity of 94.6%, AUROC 98.8%). Artificial intelligence models have potential to become valuable tools for predicting ICP insults in advance during neurointensive care. The fact that common off-the-shelf models, such as XGBoost, performed well in predicting ICP insults opens new possibilities that can lead to faster advances in the field and earlier clinical implementations.

Quantitative assessment of population risk to tropical cyclones using hybrid modeling combining GAM and XGBoost: A case study of Hainan Province

Global climate change is expected to increase the proportion of intense tropical cyclones (TCs) in the Northwest Pacific. This study focused on how factors, especially extreme events, may affect disaster losses. To address this issue, an event-based multivariate TC risk assessment model, which employs copulas, a generalized additive model, and undersampling extreme gradient boosting decision tree techniques, is developed to enhance the accuracy of disaster loss prediction. The results suggest that on Hainan Island, the rate of the affected population is positively correlated with the maximum wind speed and maximum daily rainfall while negatively correlated with the gross domestic product and elevation. The study also shows that the TC risk in the cities of Hainan Island increases as the return period expands, and each return period scenario shows a unique geospatial distribution of TC risk on Hainan Island, with higher risks in the coastal and eastern regions. These results highlight the importance of implementing effective disaster management strategies to mitigate the impact of severe TCs in the region.

Wheat Yield Prediction Using Machine Learning Method Based on UAV Remote Sensing Data

Accurate forecasting of crop yields holds paramount importance in guiding decision-making processes related to breeding efforts. Despite significant advancements in crop yield forecasting, existing methods often struggle with integrating diverse sensor data and achieving high prediction accuracy under varying environmental conditions. This study focused on the application of multi-sensor data fusion and machine learning algorithms based on unmanned aerial vehicles (UAVs) in wheat yield prediction. Five machine learning (ML) algorithms, namely random forest (RF), partial least squares (PLS), ridge regression (RR), k-nearest neighbor (KNN) and extreme gradient boosting decision tree (XGboost), were utilized for multi-sensor data fusion, together with three ensemble methods including the second-level ensemble methods (stacking and feature-weighted) and the third-level ensemble method (simple average), for wheat yield prediction. The 270 wheat hybrids were used as planting materials under full and limited irrigation treatments. A cost-effective multi-sensor UAV platform, equipped with red–green–blue (RGB), multispectral (MS), and thermal infrared (TIR) sensors, was utilized to gather remote sensing data. The results revealed that the XGboost algorithm exhibited outstanding performance in multi-sensor data fusion, with the RGB + MS + Texture + TIR combination demonstrating the highest fusion performance (R2 = 0.660, RMSE = 0.754). Compared with the single ML model, the employment of three ensemble methods significantly enhanced the accuracy of wheat yield prediction. Notably, the third-layer simple average ensemble method demonstrated superior performance (R2 = 0.733, RMSE = 0.668 t ha−1). It significantly outperformed both the second-layer ensemble methods of stacking (R2 = 0.668, RMSE = 0.673 t ha−1) and feature-weighted (R2 = 0.667, RMSE = 0.674 t ha−1), thereby exhibiting superior predictive capabilities. This finding highlighted the third-layer ensemble method’s ability to enhance predictive capabilities and refined the accuracy of wheat yield prediction through simple average ensemble learning, offering a novel perspective for crop yield prediction and breeding selection.

Predicting the risk of primary Sjögren's syndrome with key N7-methylguanosine-related genes: A novel XGBoost model

ObjectivesN7-methylguanosine (m7G) plays a crucial role in mRNA metabolism and other biological processes. However, its regulators' function in Primary Sjögren's Syndrome (PSS) remains enigmatic. MethodsWe screened five key m7G-related genes across multiple datasets, leveraging statistical and machine learning computations. Based on these genes, we developed a prediction model employing the extreme gradient boosting decision tree (XGBoost) method to assess PSS risk. Immune infiltration in PSS samples was analyzed using the ssGSEA method, revealing the immune landscape of PSS patients. ResultsThe XGBoost model exhibited high accuracy, AUC, sensitivity, and specificity in both training, test sets and extra-test set. The decision curve confirmed its clinical utility. Our findings suggest that m7G methylation might contribute to PSS pathogenesis through immune modulation. Conclusionsm7G regulators play an important role in the development of PSS. Our study of m7G-realted genes may inform future immunotherapy strategies for PSS.

Specific Antibiotics Increase the Risk of Flare-Ups in Patients with Inflammatory Bowel Disease: Results from a Danish Nationwide Population-Based Nested Case-Control Study.

Inflammatory bowel disease [IBD] patients have a relapsing-remitting disease course, and amongst environmental factors that aggravate the disease course, common drugs aside from non-steroidal anti-inflammatory drugs have not been studied in detail. While the microbiome is considered to play a significant role on the disease course, the impact of antibiotics is poorly understood. This study investigated the potential impact of different classes of antibiotics on the course of disease in IBD using the Danish National Patient Registry. Danish IBD patients were studied using two nested case-control cohorts exploring associations between antibiotic types and IBD flare-ups, defined as IBD-related hospitalizations and/or high-dose systemic steroid exposure. Multivariate logistic regression and eXtreme Gradient Boosted decision tree [GBDT] machine learning methods evaluated antibiotic risks. Two cohorts with 15 636 and 5178 patients were analysed for risk of hospitalization and course of steroids, respectively. The risk of a flare-up was significantly increased with antecedent exposure to quinolones (ATC:J01M; odds ratio [OR]: 3.04-3.82), antimycotics [ATC:J02A; OR: 1.50-2.30], agents against amoebiasis and protozoal infections [ATC:P01A; OR: 1.95-3.18], intestinal anti-infectives [ATC:A07A; OR: 2.09-2.32], and beta-lactam antibiotics [ATC:J01C; OR: 1.36]. The GBDT models achieved an area under the curve of 0.71-0.85 for predicting flare-ups, with the same above-mentioned antibiotics being in the ten most important variables. We found distinctive antibiotics to be significantly associated with an increased risk of IBD flare-ups. Our findings are corroborated by our GBDT machine learning models. Healthcare providers should be aware of the deleterious potential of specific antibiotic groups in patients with IBD only using these agents in a restrictive manner or preferentially consider alternative antibiotic groups.

Intrusion detection system for vehicular ad hoc network attacks based on machine learning techniques

ABSTRACT The vehicular ad-hoc network (VANET) is a new subcategory of mobile ad-hoc networks that has excellent potential application in intelligent transportation systems. The cyberattack on vehicles can lead to privacy leaks and financial losses and could even become a national public safety issue. In this paper, a multi-level Intrusion Detection System based on Machine Learning (ML) algorithms is developed in order to detect multiple types of attacks on VANET. The proposed model combines two feature selection techniques: random forest (RF) and fast correlation-based filter (FCBF) and four tree-based algorithms: extreme gradient boosting decision tree, RF, decision tree, and ExtraTrees, for classification performance. Moreover, the model is evaluated using CICIDS2017 dataset and Python ML libraries in Jupyter Notebook such as scikit-learn and Pandas. Experiment results show that the proposed model using the stacking method achieves 99.86% attack detection accuracy, 99.85% precision with hypo-parameter (HPO), and 99.83% attack detection accuracy without using HPO. The experimental findings on the vehicular network demonstrate the feasibility of the proposed solution and the ability to implement it in real-time environments.

Beyond Lexical Boundaries: LLM-Generated Text Detection for Romanian Digital Libraries

Machine-generated content reshapes the landscape of digital information; hence, ensuring the authenticity of texts within digital libraries has become a paramount concern. This work introduces a corpus of approximately 60 k Romanian documents, including human-written samples as well as generated texts using six distinct Large Language Models (LLMs) and three different generation methods. Our robust experimental dataset covers five domains, namely books, news, legal, medical, and scientific publications. The exploratory text analysis revealed differences between human-authored and artificially generated texts, exposing the intricacies of lexical diversity and textual complexity. Since Romanian is a less-resourced language requiring dedicated detectors on which out-of-the-box solutions do not work, this paper introduces two techniques for discerning machine-generated texts. The first method leverages a Transformer-based model to categorize texts as human or machine-generated, while the second method extracts and examines linguistic features, such as identifying the top textual complexity indices via Kruskal–Wallis mean rank and computes burstiness, which are further fed into a machine-learning model leveraging an extreme gradient-boosting decision tree. The methods show competitive performance, with the first technique’s results outperforming the second one in two out of five domains, reaching an F1 score of 0.96. Our study also includes a text similarity analysis between human-authored and artificially generated texts, coupled with a SHAP analysis to understand which linguistic features contribute more to the classifier’s decision.

Dynamic bond stress-slip relationship of steel reinforcing bars in concrete based on XGBoost algorithm

The bond stress-slip relationship between a deformed steel bar and concrete under various loading rates is required to appropriately simulate the ultimate conditions of concrete structures under dynamic loading. Furthermore, the corresponding key bond parameters are crucial to ensure the accuracy of the bond-slip model. This study analyzed the effect of concrete strength, geometric profile of steel bar, concrete cover, and stress state around the bond region on the dynamic bond parameters and bond-slip curves. For this purpose, a database containing 1056 pullout specimens was established, which was used to train three ensemble learning models, including Extreme Gradient Boosting Decision Tree (XGBoost), Gradient Boosting Regression Tree, and Random Forest, thereby predicting the bond parameters. The ensemble learning models exhibited higher accuracy than existing empirical models, with the XGBoost algorithm attaining the highest accuracy. Furthermore, the predicted bond parameters were found to express the bond response of the steel bar under different loading rates and lateral confinements with good accuracy. The model proposed in this study provides a new approach for determining the bond parameters and improves the accuracy of bond stress-slip model.

An Intrusion Detection Model Based on Feature Selection and Improved One-Dimensional Convolutional Neural Network

The problem of intrusion detection has new solutions, thanks to the widespread use of machine learning in the field of network security, but it still has a few issues at this time. Traditional machine learning techniques to intrusion detection rely on expert experience to choose features, and deep learning approaches have a low detection efficiency. In this paper, an intrusion detection model based on feature selection and improved one-dimensional convolutional neural network was proposed. This model first used the extreme gradient boosting decision tree (XGboost) algorithm to sort the preprocessed data, and then it used comparison to weed out 55 features with a higher contribution. Then, the extracted features were fed into the improved one-dimensional convolutional neural network (I1DCNN), and this network training was used to complete the final classification task. The feature selection and improved one-dimensional convolutional neural network (FS-I1DCNN) intrusion detection model not only solved the traditional machine learning method of relying on expert experience to extract features but also improved the detection efficiency of the model, reduced the training time while reducing the dimension, and increased the overall accuracy. In comparison to the I1DCNN model without feature extraction and the conventional one-dimensional convolutional neural network (1DCNN) model, the experimental results demonstrate that the FS-I1DCNN model’s overall accuracy increases by 0.67% and 2.94%, respectively. Its accuracy, precision, recall, and F1-score were significantly better than those of the other intrusion detection models, including SVM and DBN.

The use of machine learning to predict somatic cell count status in dairy cows post-calving.

Udder health remains a priority for the global dairy industry to reduce pain, economic losses, and antibiotic usage. The dry period is a critical time for the prevention of new intra-mammary infections and it provides a point for curing existing intra-mammary infections. Given the wealth of udder health data commonly generated through routine milk recording and the importance of udder health to the productivity and longevity of individual cows, an opportunity exists to extract greater value from cow-level data to undertake risk-based decision-making. The aim of this research was to construct a machine learning model, using routinely collected farm data, to make probabilistic predictions at drying off for an individual cow's risk of a raised somatic cell count (hence intra-mammary infection) post-calving. Anonymized data were obtained as a large convenience sample from 108 UK dairy herds that undertook regular milk recording. The outcome measure evaluated was the presence of a raised somatic cell count in the 30 days post-calving in this observational study. Using a 56-farm training dataset, machine learning analysis was performed using the extreme gradient boosting decision tree algorithm, XGBoost. External validation was undertaken on a separate 28-farm test dataset. Statistical assessment to evaluate model performance using the external dataset returned calibration plots, a Scaled Brier Score of 0.095, and a Mean Absolute Calibration Error of 0.009. Test dataset model calibration performance indicated that the probability of a raised somatic cell count post-calving was well differentiated across probabilities to allow an end user to apply group-level risk decisions. Herd-level new intra-mammary infection rate during the dry period was a key driver of the probability that a cow had a raised SCC post-calving, highlighting the importance of optimizing environmental hygiene conditions. In conclusion, this research has determined that probabilistic classification of the risk of a raised SCC in the 30 days post-calving is achievable with a high degree of certainty, using routinely collected data. These predicted probabilities provide the opportunity for farmers to undertake risk decision-making by grouping cows based on their probabilities and optimizing management strategies for individual cows immediately after calving, according to their likelihood of intra-mammary infection.

Abstract 17213: Machine Learning-Based Prediction of Type A Aortic Dissection

Background: Existing risk predictors of aortic dissection have certain limitations. We hypothesized that machine learning models trained on clinical, demographic, and anthropometric features can further improve the prediction of patient outcomes. Objective: This study aims to develop a machine learning model that predicts type A aortic dissection and can help clinical decision making. Methods: This cohort study used the Yale Aortic Institute database. The models incorporated variables spanning demographic, anthropometric, medical history, radiological, and laboratory domains. The models were trained and validated using stratified 10-fold cross-validation. Hyperparameters for each algorithm were tuned through grid-search on the training folds. The models were trained to optimize the area under the receiver operator characteristic curve (AUROC) and were assessed in a held-out test set. Results: A total of 2,109 patients were analyzed in our study. Among them, 271 were diagnosed with type A aortic dissection. The models demonstrated strong performance on the held-out test set. Specifically, the extreme gradient boosting decision tree model achieved an AUROC of 0.821, while the random forest model achieved an AUROC of 0.820. Importantly, these models outperformed the prediction of type A aortic dissection when based solely on the ascending aorta diameter, which had an AUROC of 0.549. Besides the ascending aorta diameter, the key predictors were age, weight, height, family history, smoking, bicuspid aortic valve, and hypertension. Conclusion: We developed a machine learning model that provides an individualized prediction of the development of type A aortic dissection. This approach provides an accessible, efficient, and remote tool to identify high-risk patients.

A Machine Learning Approach for Physical Activity Recognition in Cystic Fibrosis

ABSTRACT This study aimed to develop and validate machine learning models to predict intensities in children and adolescents with cystic fibrosis (CF) across different accelerometry brands and placements. Thirty-five children and adolescents with CF (11.6 ± 2.8 yrs; 15 girls) and 28 healthy youth (12.2 ± 2.7 yrs; 16 girls) performed six activities whilst wearing GENEActivs (both wrists) and ActiGraphs GT9X (both wrists and waist). Three supervised learning classifiers (K-Nearest Neighbour, Random Forest and eXtreme Gradient Boosted Decision Tree) were used to identify the input signal pattern for each PA type and intensity, with a 10-fold cross-validation utilized to assess the performance of the classifiers. ActiGraph GT9X on the dominant wrist and waist and GENEActiv on the dominant wrist failed to predict vigorous intensity PA activities. All other models, for activity type and intensities, exceeded 97% accuracy, with a sensitivity and specificity of greater than 95%, irrespective of accelerometer brand, placement or health condition.

Career path prediction using XGBoost Model and students’ academic results

This paper proposes an approach for constructing a system for career prediction by applying the eXtreme Gradient Boosting (XGBoost) Decision Tree model to the academic results of Ho Chi Minh International University’s School of Computer Science and Engineering graduates in the past 5 years. Initially, the dataset is cleaned up and normalized to be usable for the prediction algorithm with the help of Python 3 programming language. It is then split into 2 subsets: one for training (80 percent) and the other for testing (20 percent). After that, the algorithm uses the training subset to build the classification model. Finally, the testing subset is loaded into the model to predict each student’s career path based on the respective inputs and hyper-parameters tuning is employed to boost the model’s accuracy. By utilizing this solution, the problem related to predicting students’ future career paths based on their performance throughout their years studying at the university can be adequately addressed and handled.

The Influence of Transportation Accessibility on Traffic Volumes in South Korea: An Extreme Gradient Boosting Approach

This study explored how transportation accessibility and traffic volumes for automobiles, buses, and trucks are related. This study employed machine learning techniques, specifically the extreme gradient boosting decision tree model (XGB) and Shapley Values (SHAP), with national data sources in South Korea collected from the Korea Transport Institute, Statistics Korea, and National Spatial Data Infrastructure Portal. Several key findings of feature importance and plots in non-linear relationships are as follows: First, accessibility indicators exhibited around 5 to 10% of feature importance except for Mart (around 50%). Second, better accessibility to public transportation infrastructures, such as bus stops and transit stations, was associated with higher annual average daily traffic (AADT), particularly in metropolitan areas including Seoul and Busan. Third, access to large-scale markets may have unintended effects on traffic volumes for both vehicles and automobiles. Fourth, it was shown that lower rates of AADT were associated with higher accessibility to elementary schools for all three modes of transportation. This study contributes to (1) understanding complex relationships between the variables, (2) emphasizing the role of transportation accessibility in transportation plans and policies, and (3) offering relevant policy implications.

Coupling of XGBoost ensemble methods and discrete element modelling in predicting autogenous grinding mill throughput

Autogenous Grinding (AG) is one the most important mineral processing assets in the comminution circuit, and its production performance directly impacts on the circuit throughput. Understanding the key parameters impacting the AG mill throughput is critical to reducing the operational variability and to improve throughput predictions. This study aims to develop a coupled discrete element modelling and machine learning ensemble method for AG mill throughput predictions. Discrete Element Modelling (DEM) was coupled to an extreme gradient boosting (XGBoost) decision tree method to construct an accurate mill throughput prediction model. Hyperparameter optimizations and model pruning were conducted to improve the generalized model prediction capability. A suite of operational data as well as DEM modelling results from a 32FT AG mill were used to train and validate the developed model. Results indicated that the coupled method showed good agreement with the validation dataset. The impact from each mill operational parameter on the mill throughput was also evaluated and ranked, from which the improvement strategy on mill operation can then be devised.

Exploration the pathways of connected electric vehicle design: A vehicle-environment cooperation energy management strategy

The advance in Internet of Vehicles (IoVs) enables an information-aggregated environment, underpinning the connected electric vehicle (cEV) development. The sensed multi-range driving condition information can consummate energy management in cEVs. IoVs based cEV design is in initial stage. The solutions that demonstrate the critical role of IoVs in optimal energy management in real time have not reach to state-of-the-art. In this study, a vehicle-environment cooperation energy management strategy (VEC-EMS) is proposed for cEV based on the explicitly framed cooperation mechanism in IoVs. First, an IoVs framework and inner cooperation mechanism are elaborated. Then, the VEC-EMS, empowered robustness to varying driving conditions in real-time optimal implementation, is designed. The adaptability to driving conditions is attained by a future vehicle status observer (FVSO), which integrates the improved radial basis function neural network (iRBF-NN) based velocity prediction and extreme gradient boosting decision tree (XGBoost) based driving condition identification. The optimality in instant energy management is accomplished via dynamic assignment of the optimized control thresholds according to the results of FVSO. The control thresholds are optimized by the improved Beetle Antennae Search (iBAS). At last, evaluation manifests that the proposed EMS can manage power flow within the electric powertrain, highlighting its anticipated preferable performance which increases by nearly 8% compared with a normal rule-based energy management strategy.

Characterization of oxidative damage induced by nanoparticles via mechanism-driven machine learning approaches

The wide application of TiO2-based engineered nanoparticles (nTiO2) inevitably led to release into aquatic ecosystems. Importantly, increasing studies have emphasized the high risks of nTiO2 to coastal environments. Bivalves, the representative benthic filter feeders in coastal zones, acted as important roles to assess and monitor the toxic effects of nanoparticles. Oxidative damage was one of the main toxic mechanisms of nTiO2 on bivalves, but the experimental variables/nanomaterial characteristics were diverse and the toxicity mechanism was complex. Therefore, it was very necessary to develop machine learning model to characterize and predict the potential toxicity. In this study, thirty-six machine learning models were built by nanodescriptors combined with six machine learning algorithms. Among them, random forest (RF) – catalase (CAT), k-neighbors classifier (KNN) - glutathione peroxidase (GPx), neural networks - multilayer perceptron (ANN) – glutathione s-transferase (GST), random forest (RF) - malondialdehyde (MDA), random forest (RF) - reactive oxygen species (ROS), and extreme gradient boosting decision tree (XGB) - superoxide dismutase (SOD) models performed good with high accuracy and balanced accuracy for both training sets and external validation sets. Furthermore, the best model revealed the predominant factors (exposure concentration, exposure periods, and exposure matrix) influencing the oxidative stress induced by nTiO2. These results showed that high exposure concentrations and short exposure-intervals tended to cause oxidative damage to bivalves. In addition, gills and digestive glands could be vulnerable to nTiO2-induced oxidative damage as tissues/organs differences were the important factors controlling MDA activity. This study provided insights into important nano-features responsible for the different indicators of oxidative stress and thereby extended the application of machine learning approaches in toxicological assessment for nanoparticles.

An improved spatial case-based reasoning considering multiple spatial drivers of geographic events and its application in landslide susceptibility mapping

Traditional case-based reasoning methods overlook non-stationary spatial drivers of geographical events such as heterogeneity, dependence, and accumulation in case representation, and directly obtain the solution of the most similar cases in case reuse instead of considering the interference of fake similar cases to eliminate the contingency of reasoning, which leads to poor interpretations and low efficiency decisions in complex and heterogeneous geographical environments. This study proposes an improved spatial case-based reasoning (SCBR) considering multiple spatial drivers to overcome above problems and uses landslide susceptibility mapping as an example. Specifically, these spatial drivers were captured, extracted, and integrated into case representation by using geographic self-organizing mapping algorithm, spatial statistic, and spatial adjacent matrix, respectively. Additionally, the K-nearest neighbor method as case retrieval was introduced to retrieve the K similar cases based on the local and global similarity reasoning. Finally, the Gaussian process regression as case reuse method was generated to landslide susceptibility index under the assumption that K similar cases follows Gaussian distribution. Our experimental results show that the precision, F1, recall, and kappa of the proposed SCBR method are 0.974, 0.976, 0.979, and 0.953 which are higher than those of the traditional case-based reasoning (0.931, 0.941, 0.953, and 0.881), long short-term memory (0.951, 0.933, 0.915, and 0.870), and extreme gradient boosting decision tree (0.963, 0.967, 0.972, and 0.945), respectively. In general, the novel approach with better predictive performance can help decision makers to develop policies that reduce the loess of landslides and apply to similar geological events.

Equity in Active and Shared Transportation: An Investigation of Barriers and Individual and Contextual Factors

To achieve the goal of encouraging more people to use sustainable modes of transportation, thereby reducing carbon emissions, it is necessary to understand the equitable distribution of active and shared transportation among demographic groups. This study investigated which individual and contextual factors influence perceptions of barriers to public transit, micromobility, and carpool. A non-parametric method (extreme gradient boosting decision tree) was employed to accurately model these relationships and rank the importance of the factors. To enhance model interpretability, Shapley Additive Explanations (SHAP) was employed to explain the model outcome. Three metropolitan U.S. regions—Greater Los Angeles, Greater Houston, and Virginia and Washington D.C.—were selected as the application regions for this developed methodology. Based on the findings, it was found that people who drive for most of their trips are more inclined to perceive the reliability of public transit as unsatisfactory and have a lack of familiarity with micromobility. Women are more prone to perceive public transit as unsafe, uncomfortable, and inaccessible, and minority groups are least likely to be unsatisfied with public transit, suggesting that they might be more willing to endure inadequate transportation service. Thus, ensuring that these groups feel that they have the right and ability to speak up about their experience and concerns is critical. This research provides insights for transportation agencies to develop equitable improvement and communication strategies for active and shared transportation systems, which is imperative for the widespread adoption of sustainable modes for all people regardless of demographic characteristics.