Neighbors In Network Research Articles

Integrating machine learning for treatment decisions in anterior open bite orthodontic cases: A retrospective study.

This article explores the integration of machine learning (ML) algorithms to aid in treatment planning and extraction decisions for anterior open bite cases, leveraging demographic, clinical, and radiographic data to predict treatment outcomes and informed decision-making. A retrospective study was conducted using patient data from the University of Illinois Chicago Department of Orthodontics. Data included demographic, clinical, and radiographic information from 115 anterior open bite patients who successfully completed their treatment. ML algorithms, including random forest, support vector machine, k-nearest neighbor, and convolutional neural networks (CNN), were trained on a subset of the data to predict treatment outcomes. Significant differences were observed in the percentages of males and females between the extraction and nonextraction groups and cephalometric variables between the two groups, which include maxillary depth, maxillary height, SN-palatal plane, facial angle, facial axis-Ricketts, FMA, total facial height, lower facial height, SNA, SNB, and SN-MP e ML algorithms examined consisted of CNN2, CNN1, and Random Forest, which demonstrated the highest accuracy rates (∼83%), while k-Nearest Neighbor had the lowest (∼73%). Key features influencing accuracy included crowding, SN-palatal plane, SNA, FMA, molar relation, and facial height measurements. The study's evaluation of AI algorithms showed that CNN2, CNN1, and random forest had an accuracy of approximately 83% in classifying extraction versus nonextraction cases. Notably, features such as U-crowding, L-crowding, SN-palatal plane, SNA, FMA, molar relation, total facial height, lower facial height, and facial axis-Ricketts were most influential in achieving accuracy rates comparable to traditional methods.

Proximity proteomics provides a new resource for exploring the function of Afadin and the complexity of cell-cell adherens junctions.

The network of proteins at the interface between cell-cell adherens junctions and the actomyosin cytoskeleton provides robust yet dynamic connections that facilitate cell shape change and motility. While this was initially thought to be a simple linear connection via classic cadherins and their associated catenins, we now have come to appreciate that many more proteins are involved, providing robustness and mechanosensitivity. Defining the full set of proteins in this network remains a key objective in our field. Proximity proteomics provides a means to define these networks. Mammalian Afadin and its Drosophila homolog Canoe are key parts of this protein network, facilitating diverse cell shape changes during gastrulation and other events of embryonic morphogenesis. Here we report results of several proximity proteomics screens, defining proteins in the neighborhood of both the N- and C-termini of mammalian Afadin in the premier epithelial model, MDCK cells. We compare our results with previous screens done in other cell types, and with proximity proteomics efforts with other junctional proteins. These reveal the value of multiple screens in defining the full network of neighbors and offer interesting insights into the overlap in protein composition between different epithelial cell junctions.

Efficient Estimation for Longitudinal Networks via Adaptive Merging

Longitudinal networks consist of sequences of temporal edges among multiple nodes, where the temporal edges are observed in real-time. They have become ubiquitous with the rise of online social platforms and e-commerce, but largely under-investigated in the literature. In this paper, we propose an efficient estimation framework for longitudinal networks, leveraging strengths of adaptive network merging, tensor decomposition, and point processes. It merges neighboring sparse networks so as to enlarge the number of observed edges and reduce estimation variance, whereas the estimation bias introduced by network merging is controlled by exploiting local temporal structures for adaptive network neighborhood. A projected gradient descent algorithm is proposed to facilitate estimation, where the upper bound of the estimation error in each iteration is established. A thorough analysis is conducted to quantify the asymptotic behavior of the proposed method, which shows that it can significantly reduce the estimation error and also provides a guideline for network merging under various scenarios. We further demonstrate the advantage of the proposed method through extensive numerical experiments on synthetic datasets and a militarized interstate dispute dataset.

Classification of NSCLC subtypes using lung microbiome from resected tissue based on machine learning methods

Classification of adenocarcinoma (AC) and squamous cell carcinoma (SCC) poses significant challenges for cytopathologists, often necessitating clinical tests and biopsies that delay treatment initiation. To address this, we developed a machine learning-based approach utilizing resected lung-tissue microbiome of AC and SCC patients for subtype classification. Differentially enriched taxa were identified using LEfSe, revealing ten potential microbial markers. Linear discriminant analysis (LDA) was subsequently applied to enhance inter-class separability. Next, benchmarking was performed across six different supervised-classification algorithms viz. logistic-regression, naïve-bayes, random-forest, extreme-gradient-boost (XGBoost), k-nearest neighbor, and deep neural network. Noteworthy, XGBoost, with an accuracy of 76.25%, and AUROC (area-under-receiver-operating-characteristic) of 0.81 with 69% specificity and 76% sensitivity, outperform the other five classification algorithms using LDA-transformed features. Validation on an independent dataset confirmed its robustness with an AUROC of 0.71, with minimal false positives and negatives. This study is the first to classify AC and SCC subtypes using lung-tissue microbiome.

Assessment of using transfer learning with different classifiers in hypodontia diagnosis

BackgroundHypodontia is the absence of one or more teeth in the primary or permanent dentition during development, and radiographic imaging is the most common method of diagnosis. However, in recent years, artificial intelligence-based decision support systems have been employed to make highly accurate diagnoses. The aim of this study was to classify single premolar agenesis, multiple premolar agenesis, and without tooth agenesis using various artificial intelligence approaches.MethodsOne thousand sixty-eight panoramic radiographs from pediatric patients aged between 6 and 12 years without systemic disease were sorted into three separate classes: single premolar agenesis (n = 336), multiple premolar agenesis (n = 324), and without tooth agenesis (n = 408). Pretrained convolutional neural network models (AlexNet, DarkNet-19, DarkNet-53, DenseNet-201, EfficientNet, GoogLeNet, InceptionV3, IncResV2, MobileNetV2, NasNet-Mobile, Places365, ResNet-18, ResNet-50, ResNet-101, ShuffleNet, SqueezeNet, VGG-16, VGG-19, and Xception) were used for training with the fine-tuning method and different machine learning classifiers (decision trees, discriminant analysis, logistic regression, naive Bayes, support vector machines, nearest neighbor, ensemble method, and artificial neural network). The dataset was divided into 80% for training and 20% for testing. Performance was evaluated via accuracy, recall, precision, F1-score, specificity and area under the curve (AUC) parameters.ResultsAll of the data were classified via a VGG-19 model with a bilayered neural network classifier, which achieved 95.63% accuracy, 93.26% precision, 93.34% recall, 96.73% specificity, 93.25% F1-score and 95.03% AUC and was identified as the most successful model. The accuracy values for this model were distributed as follows: 96.72% for without tooth agenesis, 95.79% for multiple premolar agenesis, and 94.39% for single premolar agenesis.ConclusionsSuccessful results of pretrained models have been demonstrated for the radiographic diagnosis of hypodontia in pediatric patients. It is expected that artificial intelligence approaches will facilitate the diagnosis of hypodontia.

Exploring the Role of Communication Asset Mapping (CAM) as a Strategy to Promote Hereditary Cancer Risk Assessment Information Within African American Communities.

Objective: African Americans (AAs) carry the largest burden for almost every type of cancer in the US and are also more likely to die from cancer. Approximately 10% of cancers can be explained by a hereditary factor and detected earlier. Many AAs, however, have inequitable access to hereditary cancer risk assessment (HCRA) tools and information, further exacerbating disparities in cancer rates. Innovative communication strategies to promote community-based HCRA information have promise as a means encouraging optimal primary cancer screening among AAs. The current pilot study followed a participatory process where researchers engaged with a Community Advisory Board (CAB) to explore how Communication Asset Mapping (CAM) could assist lay health advisors with the dissemination of evidence-based HC/RA information within AA faith communities. Methods: The research team and CAB conducted exploratory community-engaged group discussions with residents (n = 21) guided by Communication Infrastructure Theory, and used a community-engaged mapping process to inform the development of a CAM dissemination strategy. Results: Through textual analysis, the following conclusions were reached: (1) optimal locations (e.g., community centers) within specified neighborhood networks should have representatives who are trusted ambassadors to assist with HCRA information dissemination; (2) trusted community member voices should fully represent the neighborhood network in the community-engagement mapping process; (3) well-known and frequented geographic locations should provide a true representation of participants' neighborhoods to create a robust health information network concerning HCRA. Conclusions: Community residents appreciated the engagement process; however, they felt that its impact was limited due to the lack of community voices within their neighborhoods to identify important communication resources within the network for optimal HCRA information dissemination. CAM, therefore, is an important public health strategy for the identification of trusted networks and useful communication resources within these networks. The strategy was also helpful in pinpointing people who could be critical communicators of emerging health information akin to HCRA.

Integrating machine learning and nano-QSAR models to predict the oxidative stress potential caused by single and mixed carbon nanomaterials in algal cells.

In silico methods are increasingly important in predicting the ecotoxicity of engineered nanomaterials (ENMs), encompassing both individual and mixture toxicity predictions. It is widely recognized that ENMs trigger oxidative stress effects by generating intracellular reactive oxygen species (ROS), serving as a key mechanism in their cytotoxicity studies. However, existing in silico methods still face significant challenges in predicting the oxidative stress effects induced by ENMs. Herein, we utilized laboratory-derived toxicity data and machine learning methods to develop quantitative nanostructure-activity relationship (nano-QSAR) classification and regression models, aiming to predict the oxidative stress effects of five carbon nanomaterials (fullerene, graphene, graphene oxide, single-walled carbon nanotubes, and multi-walled carbon nanotubes) and their binary mixtures on Scenedesmus obliquus cells. We constructed five nano-QSAR classification models by combining zeta potential (ζP) with the C4.5 decision tree, support vector machine, artificial neural network, naive Bayes, and K-nearest neighbor algorithms. Moreover, we constructed three classification models by integrating the features including ζP, hydrodynamic diameter (DH), and specific surface area (SSA) with the logistic regression, random forest, and Adaboost algorithms. The Accuracy, Recall, Precision and harmonic mean of Precision and Recall (F1-score) values of these models were all higher than 0.600, indicating an excellent performance in distinguishing whether CNMs have the potential to generate ROS. In addition, using the ζP, DH, and SSA descriptors, we combined decision tree regression, random forest regression, gradient boosting, and the Adaboost algorithm, and successfully constructed four nano-QSAR regression models with applicable application domains (all training and testing data points lie within 95% confidence intervals), goodness-of-fit (Rtrain2 ≥ 0.850), and robustness (cross-validation R2 ≥ 0.650) as well as predictive power (Rtest2 ≥ 0.610). The method developed would establish a fundamental basis for more precise evaluations of ecological risks posed by these materials from a mechanistic standpoint.

Compiler Sequence Optimization Using Machine Learning Prediction Method

Compiler optimization is crucial in improving program performance by improving execution speed, reducing memory usage, and minimizing energy consumption. Nevertheless, modern compilers, such as LLVM, with their numerous optimization passes, present a significant challenge in identifying the most effective sequence for optimizing a program. This study addresses the complex problem of determining optimal compiler optimization sequences within the LLVM framework, which encompasses 64 optimization passes, causing in an immense search space of 264264. Identifying the ideal sequence for even simple code can be an arduous task, as the interactions between passes are intricate and unpredictable. The primary objective of this research is to utilize machine-learning techniques to predict effective optimization sequences that outperform the default -O2 and -O3 optimization flags. The methodology involves generating 2,000 sequences per program and picking the one that achieves the shortest execution time. Three machine learning models—K-Nearest Neighbor (KNN), Decision Tree (DT), and Feedforward Neural Network (FFNN)—were employed to predict the optimization sequences based on features extracted from programs during execution. The study used benchmarks from Polybench, Shootout, and Stanford suites, each with varying problem sizes, to validate the proposed technique. The results demonstrate that the KNN model produced optimization sequences with superior performance compared to DT and FFNN. On average, KNN achieved execution times that were 2.5 times faster than those achieved using the O3 optimization flag. This research contributes to the field by programming the process of selecting optimal compiler sequences, which significantly reduces execution time and eliminates the need for manual tuning. It highlights the potential of machine learning in compiler optimization, offering a robust and scalable approach to improving program performance and setting the foundation for future advancements in the domain.

Prediction of induction motor faults using machine learning.

Unplanned downtime in industrial sectors presents significant challenges, impacting both production efficiency and profitability. To tackle this issue, companies are actively working towards optimizing their operations and reducing disruptions that hinder their ability to meet customer demands and financial goals. Predictive maintenance, utilizing advanced technologies like data analytics, machine learning, and IoT devices, offers real-time equipment data monitoring and analysis. This research study centers on the development of a versatile machine-learning model for predicting faults in induction motors within industrial environments. Implementing such a model can enable proactive maintenance, ultimately leading to decreased downtime in industrial operations. The study involved the acquisition of a dataset comprising healthy and faulty conditions of four 3-phase induction motors, along with relevant features for fault prediction. Multiple machine learning algorithms were trained using this dataset, exhibiting promising performance. The Random Forest (RF) model achieved the highest accuracy at 0.91, closely followed by the Artificial Neural Network (ANN) and k-nearest Neighbors (k-NN) models, both achieving an accuracy of 0.9. Meanwhile, the Decision Tree (DT) model showed the lowest accuracy at 0.89. Further model evaluation was carried out using a confusion matrix, which provided a detailed breakdown of the models' performance for each class, revealing the number of correctly and incorrectly classified induction motor conditions. The results from the confusion matrix indicate that the models effectively classified the various states and conditions of the induction motors. To enhance model performance in future work, potential avenues include refining the ANN and RF models, exploring transfer learning or ensemble methods, and incorporating diverse datasets to improve generalization.

Genetic diversity and population structure of Canarian chicken using microsatellite DNA markers

The Canary Islands have historically been a crossway among at least three continents; therefore, the genetic influences on their local animal breeds have been extremely diverse. In the Canarian chicken population, genetic diversity is evident but it has never been studied. The aim of this study was to assess the population structure and genetic diversity within the Canarian chicken population, as well as between the Canarian chicken population, the Spanish local populations, and the commercial chicken populations with microsatellites, with a view to reinforce the official recognition of the breed, for the design and development of a conservation program. Blood samples were collected at random from 198 animals of the Canarian and compared with Spanish local and commercial strains, in order to determine differentiation and genetic relatedness. The five phenotypic varieties of Canarian chicken, the population structure and genetic diversity within the Canarian chicken population had a higher unbiased expected heterozygosity than the observed heterozygosity. Comparing the Canarian vs other local Spanish breeds and commercial strains, FST was relatively high (0.179) (0.164 – 0.195), the neighborhood network showed that the Canarian varieties did not cluster with the other Spanish breeds. The STRUCTURE, confirmed that the Rubilana variety differed from the other four Canary Islands. We conclude that the Canary Islands chicken population shows a differentiated genetic profile, versus other Spanish and cosmopolitan breeds. The theory that the existence of genetic varieties is based on the color of feathers was definitively discarded, except in the Rubilana population, which could be admitted as a genetically different variety.

Genomic investigation of plant secondary metabolism: insights from synteny network analysis of oxidosqualene cyclase flanking genes.

The clustered distribution of genes involved in metabolic pathways within the plant genome has garnered significant attention from researchers. By comparing and analyzing changes in the flanking regions of metabolic genes across a diverse array of species, we can enhance our understanding of the formation and distribution of biosynthetic gene clusters (BGCs). In this study, we have designed a workflow that uncovers and assesses conserved positional relationships between genes in various species by using synteny neighborhood networks (SNN). This workflow is then applied to the analysis of flanking genes associated with oxidosqualene cyclases (OSCs). The method allows for the recognition and comparison of homologous blocks with unique flanking genes accompanying different subfamilies of OSCs. The examination of the flanking genes of OSCs in 122 plant species revealed multiple genes with conserved positional relationships with OSCs in angiosperms. Specifically, the earliest adjacency of OSC genes and CYP716 genes first appeared in basal eudicots, and the nonrandom occurrence of CYP716 genes in the flanking region of OSC persists across different lineages of eudicots. Our study showed the substitution of genes in the flanking region of the OSC varies across different plant lineages, and our approach facilitates the investigation of flanking gene rearrangements in the formation of OSC-related BGCs.

Hither-CMI: Prediction of circRNA-miRNA Interactions Based on a Hybrid Multimodal Network and Higher-Order Neighborhood Information via a Graph Convolutional Network.

Numerous studies show that circular RNA (circRNA) functions as a sponge for microRNA (miRNA), significantly regulating gene expression by interacting with miRNA, which in turn affects the progression of human diseases. Traditional experimental approaches for investigating circRNA-miRNA interactions (CMI) are both time-consuming and costly, making computational methods a valuable alternative. Hence, we propose a computational model for predicting CMI, leveraging a hybrid multimodal network and higher-order neighborhood information (Hither-CMI). Specifically, Hither-CMI employs Multiple Kernel Learning (MKL) to integrate sequence, structure, and expression similarity networks of circRNA and miRNA, resulting in a hybrid multimodal network. Next, an enhanced Graph Convolutional Network (GCN) is utilized to combine the circRNA-miRNA hybrid multimodal network with the CMI association network, producing a hybrid higher-order embedding representation. Finally, the XGBoost classifier is applied for training and prediction. The Hither-CMI model achieved a predicted AUC value of 0.9134. In case studies, 25 out of the top 30 predicted CMI were confirmed by recent literature. These extensive experimental results further validate the effectiveness of Hither-CMI in predicting potential CMI, making it a promising prescreening tool for further biological research.

Legal geographies in the making: Urban inequality, neighbourhood networks, and pandemic territorialities

Legal geographies in the making: Urban inequality, neighbourhood networks, and pandemic territorialities

Psoriatic arthritis in psoriasis: optimizing the current screening system for psoriatic arthritis based on serum data from U.S. and Chinese populations.

Psoriatic arthritis (PSA) is an inflammatory joint disease associated with psoriasis (PSO) that can be easily missed. Existing PSA screening tools ignore objective serologic indicators. The aim of this study was to develop a disease screening model and the Psoriatic Arthritis Inflammation Index (PSAII) based on serologic data to enhance the efficiency of PSA screening. A total of 719 PSO and PSA patients from the National Health and Nutrition Examination Survey (NHANES) (as training set and test set) and 135 PSO and PSA patients who were seen at The First Affiliated Hospital of Zhejiang Chinese Medical University (as external validation set) were selected, 31 indicators for these patients were collected as potential input features for the model. Least Absolute Shrinkage and Selection Operator (LASSO) was used to identify PSA-related features. Five models of logistic regression (LR), random forest, k-nearest neighbor, gradient augmentation and neural network were developed in the training set using quintuple cross validation. And we developed PSAII based on the results of LASSO regression and weights of logistic model parameters. All performance metrics are derived on the test set and the external validation set. Five variables were selected to build models, including age, lymphocyte percentage, neutrophil count, eosinophilic count, and C-reactive protein. In all established models, the LR model performed the best, with an Area Under Curve (AUC) of 0.87 (95% confidence interval (CI): 0.83-0.90) on the test set; on the external validation set the AUC was 0.82 (95%CI: 0.74-0.90). The PSAII formula was PSAII = percentage of lymphocytes × C-reactive protein/(neutrophil count × eosinophilic count × 10). The AUC of PSAII in the test is 0.93 (95%CI: 0.88-0.97), and the cutoff value is 18. The AUC of the external validation set is 0.81 (95%CI: 0.72-0.89). This study developed and validated five models to assist screening for PSA by analyzing serum data from NHANES and Chinese populations. The LR model demonstrated the best performance. We created PSAII for PSA screening. However, the high false positive rate of PSAII makes it necessary to combine it with other PSA screening tools when applied.

Triplet Network for One-Shot Raman Spectrum Recognition.

Raman spectroscopy is widely used for material detection due to its specificity, but its application to spectral recognition often faces limitations due to insufficient training data, unlike fields such as image recognition. Traditional machine learning or basic neural networks are commonly used, but they have limited ability to achieve high precision. We have proposed a novel approach that combines the Triplet network (TN) and K-nearest neighbor (KNN) techniques to address this issue. TN maps the Raman spectral sequences to a 128-dimensional Euclidean space to extract features, enabling the features in the new space to more accurately represent the similarities or differences between spectra, and then utilizes the KNN algorithm to perform classification tasks in this feature space. Our method exhibits superior performance in recognizing unknown Raman spectra with minimal training samples per class. We employed a handheld Raman spectrometer with an excitation wavelength of 785 nm to collect the Raman spectra of 36 samples, including 28 safe materials and eight hazardous materials. Using only one spectrum as a support set for each category, the hazardous samples were successfully distinguished from the safe samples with an accuracy of 99.6%. Additionally, our model offers adaptability without requiring exhaustive retraining when adding new prediction classes. In situations with high background fluorescence, the TN performs better in measuring the distance between spectra of the same class than traditional distance measurement methods.

Employing machine learning models to predict pregnancy termination among adolescent and young women aged 15–24 years in East Africa

Pregnancy termination is still a sensitive and continuing public health issue due to several political, economic, religious, and social concerns. This study assesses the applications of machine learning models in the prediction of pregnancy termination using data from eleven national datasets in East Africa. Nine machine learning models, namely: Random Forests (RF), Decision Tree, Logistic Regression, Support Vector Machine, eXtreme Gradient Boosting (XGB), AdaBoost, CatBoost, K-nearest neighbor, and feedforward neural network models were used to predict pregnancy termination, with six evaluation criteria utilized to compare their performance. The pooled prevalence of pregnancy termination in East Africa was found to be 4.56%. All machine learning models had an accuracy of at least 71.8% on average. The RF model provided accuracy, specificity, precision, and AUC of 92.9%, 0.87, 0.91, and 0.93, respectively. The most important variables for predicting pregnancy termination were marital status, age, parity, country of residence, age at first sexual activity, exposure to mass media, and educational attainment. These findings underscore the need for a tailored approach that considers socioeconomic and regional disparities in designing policy initiatives aimed at reducing the rate of pregnancy terminations among younger women in the region.

The role of community leadership in building community adaptive capacity to coastal hazards – Insights from neighborhood networks in Semarang, Indonesia

Coastal communities in the Global South face significant challenges due to urbanization, population growth, and climate change-induced sea level rise. This paper contributes to the study of community adaptation in these contexts by emphasizing the critical role of community leaders in developing, maintaining, and activating community adaptive capacity. Drawing on quantitative data from a household survey and qualitative interviews with community leaders in Semarang, Indonesia, this study identifies and empirically analyzes previously overlooked linkages between community leadership and adaptive capacity. Our key findings demonstrate that effective leadership significantly impacts community adaptive capacity by building social capital, organizing collective action, accessing knowledge, managing communal funds, and representing community interests. Community leaders' influence stems from their network position as brokers and their ability to mobilize external resources and knowledge. Furthermore, we develop an empirically based framework of factors determining community leadership effectiveness, emphasizing the importance of social networks. Factors such as social capital, formal authority, influence, and legitimacy determine the scope of leaders’ actions and resources and are closely connected to the structure and dynamics of social networks. Our research contributes to the literature on community adaptive capacity by underlining the critical role of community leaders and social networks in enhancing community adaptation to natural hazards, with broader implications for coastal communities across the Global South.

Synergism of Computational Simulation Technique and Machine Learning Algorithm for Prediction of Anticorrosion Properties of Some Antipyrine Derivatives.

This study aimed to predict the selected antipyrine compounds' inhibitory efficiencies and anticorrosion properties in a hydrochloric acid (HCl) environment. Molecular descriptors and input variables were obtained using density functional theory (DFT), and the variance inflation factor (VIF) was employed to reduce redundant variables, leading to the selection of seven quantum chemical descriptors as input variables. Using machine learning techniques such as K-nearest neighbor (KNN) and artificial neural network (ANN), a predictive model was built for 39 antipyrine compounds with known corrosion inhibition efficiencies for carbon and low alloy steel in hydrochloric acid solutions. The models' predictive capability was assessed using cross-validation, with the ANN model showing superior performance, achieving a coefficient of determination (R2) value of 0.715 compared to 0.548 for the KNN model. Performance metrics such as the mean square error (MSE), mean absolute error (MAE), and root-mean-square error (RMSE) further confirmed the superiority of the ANN model over the KNN model. The corrosion inhibition efficiencies (CIEs) of the selected antipyrine compounds ranged from 68.78 to 99.79%, with compound A1 demonstrating the highest CIE of 99.79% and compound A3 the lowest, as evaluated by the ANN model. Analysis of Fukui index parameters obtained from the Mulliken population analysis suggested that the nucleophilic and electrophilic sites play a crucial role in the interactions between the inhibitor and the metal atom through electron donor-acceptor interactions. Moreover, the energy of adsorption (Eads) in kcal·mol-1 decreased in the order of A1 (-187.8) > A2 (-132.0) > A2 (-84.4), with the high negative value of Eads indicating strong and spontaneous adsorption. Further analysis using radial distribution functions and molecular dynamics simulations revealed that inhibitor A1 exhibited predominantly chemisorption, inhibitor A2 showed a mixed type, and inhibitor A3 demonstrated predominantly physisorption, aligning well with the results of the predictive studies.

Tool Wear Prediction in Machining of Aluminum Matrix Composites with the Use of Machine Learning Models.

This paper discusses the diagnostic models of tool wear during face milling of Aluminum Matrix Composite (AMC), classified as a difficult-to-cut material. Prediction and classification models were considered. The models were based on one-dimensional simple regression or on multidimensional regression trees, random forest, nearest neighbor and multilayer perceptron neural networks. Measures of diagnostic signals obtained from measurements of cutting forces and vibration accelerations of the workpiece were used. The study demonstrated that multidimensional models outperformed one-dimensional models in terms of prediction accuracy and classification performance. Specifically, multidimensional predictive models exhibited lower maximum and average absolute prediction errors (0.036 mm vs. 0.050 mm and 0.026 mm vs. 0.045 mm, respectively), and classification models recorded fewer Type I and Type II errors. Despite the increased complexity, the higher predictive accuracy (up to 0.97) achieved with multidimensional models was shown to be suitable for industrial applications. However, simpler one-dimensional models offered the ad-vantage of greater reliability in signal acquisition and processing. It was also highlighted that the advantage of simple models from a practical point of view is the reduced complexity and consequent greater reliability of the system for acquiring and processing diagnostic signals.

Advancing Coastal Flood Risk Prediction Utilizing a GeoAI Approach by Considering Mangroves as an Eco-DRR Strategy

Traditional coastal flood risk prediction often overlooks critical geographic features, underscoring the need for accurate risk prediction in coastal cities to ensure resilience. This study enhances the prediction of coastal flood occurrence by utilizing the Geospatial Artificial Intelligence (GeoAI) approach. This approach employed models—random forest (RF), k-nearest neighbor (kNN), and artificial neural networks (ANN)—and compared them to the IPCC risk framework. This study used El Salvador as a demonstration case. The models incorporated seven input variables: extreme sea level, coastline proximity, elevation, slope, mangrove distance, population, and settlement type. With a recall score of 0.67 and precision of 0.86, the RF model outperformed the other models and the IPCC approach, which could avoid imbalanced datasets and standard scaler issues. The RF model improved the reliability of flood risk assessments by reducing false negatives. Based on the RF model output, scenario analysis predicted a significant increase in flood occurrences by 2100, mainly under RCP8.5 with SSP5. The study also highlights that the continuous mangrove along the coastline will reduce coastal flood occurrences. The GeoAI approach results suggest its potential for coastal flood risk management, emphasizing the need to integrate natural defenses, such as mangroves, for coastal resilience.