Machine Learning Applications Research Articles

Applying machine learning algorithms to group the energy performance of existing buildings by different granularity level of energy data

ABSTRACT Recent technology advancement has extended smart-metering capability, enabling measurement of fine-grained building energy consumption. These massive energy data provide options to apply machinelearning in building energy analysis. However, the energy details needed depend on the building feature being analyzed. Unlike system faults and user-related inefficiencies, degradation in building thermal performance results in a steady energy increase that can be reflected even with coarse-grained energy data. This study assesses whether energy data of low temporal resolution can be used to provide insight into building energy efficiency. It evaluates the application of machine learning in clustering buildings based on their thermal characteristics using energy data of different time intervals. Through k-means clustering, buildings are clustered using their hourly, daily, and monthly heating energy. The results indicate that all three granularities can identify buildings with high window transmittances and infiltration rates. These findings present an opportunity for applying machine learning not only for modern buildings with advanced metering systems but also for aging buildings that makes the majority of the existing building stock. The study presents a practical method to identify and prioritize buildings for energy retrofit, an important step toward an effective energy upgrade of a large-scale building using machine learning.

BHT-QAOA: The Generalization of Quantum Approximate Optimization Algorithm to Solve Arbitrary Boolean Problems as Hamiltonians.

A new methodology is introduced to solve classical Boolean problems as Hamiltonians, using the quantum approximate optimization algorithm (QAOA). This methodology is termed the "Boolean-Hamiltonians Transform for QAOA" (BHT-QAOA). Because a great deal of research and studies are mainly focused on solving combinatorial optimization problems using QAOA, the BHT-QAOA adds an additional capability to QAOA to find all optimized approximated solutions for Boolean problems, by transforming such problems from Boolean oracles (in different structures) into Phase oracles, and then into the Hamiltonians of QAOA. From such a transformation, we noticed that the total utilized numbers of qubits and quantum gates are dramatically minimized for the generated Hamiltonians of QAOA. In this article, arbitrary Boolean problems are examined by successfully solving them with our BHT-QAOA, using different structures based on various logic synthesis methods, an IBM quantum computer, and a classical optimization minimizer. Accordingly, the BHT-QAOA will provide broad opportunities to solve many classical Boolean-based problems as Hamiltonians, for the practical engineering applications of several algorithms, digital synthesizers, robotics, and machine learning, just to name a few, in the hybrid classical-quantum domain.

Machine learning and artificial intelligence: Enabling the clinical translation of atomic force microscopy-based biomarkers for cancer diagnosis

The influence of biomechanics on cell function has become increasingly defined over recent years. Biomechanical changes are known to affect oncogenesis; however, these effects are not yet fully understood. Atomic force microscopy (AFM) is the gold standard method for measuring tissue mechanics on the micro- or nano-scale. Due to its complexity, however, AFM has yet to become integrated in routine clinical diagnosis. Artificial intelligence (AI) and machine learning (ML) have the potential to make AFM more accessible, principally through automation of analysis. In this review, AFM and its use for the assessment of cell and tissue mechanics in cancer is described. Research relating to the application of artificial intelligence and machine learning in the analysis of AFM topography and force spectroscopy of cancer tissue and cells are reviewed. The application of machine learning and artificial intelligence to AFM has the potential to enable the widespread use of nanoscale morphologic and biomechanical features as diagnostic and prognostic biomarkers in cancer treatment.

Application of machine learning in ultrasonic pretreatment of sewage sludge: Prediction and optimization

In this research, typical industrial scenarios were analyzed optimized by machine learning algorithms, which fills the gap of massive data and industrial requirements in ultrasonic sludge treatment. Principal component analysis showed that the ultrasonic density and ultrasonic time were positively correlated with soluble chemical oxygen demand (SCOD), total nitrogen (TN), and total phosphorus (TP). Within five machine learning models, the best model for SCOD prediction was XG-boost (R2 = 0.855), while RF was the best for TN and TP (R2 = 0.974 and 0.957, respectively). In addition, SHAP indicated that the importance feature for SCOD, TN, and TP was ultrasonic time, and sludge concentration, respectively. Finally, the typical industrial scenario of ultrasonic pretreatment of sludge was analyzed. In the secondary sludge, treatment volume at 0.6 L, the pH at 7.0, and the ultrasonic time at 20 min was best to improve the SCOD. In the ultrasonic pretreatment primary sludge, treatment volume of 0.3 L, pH of 7.0, and ultrasonic time of 15 min was best to improve the SCOD. Furthermore, the ultrasonic power at 700 W and ultrasonic time at 20 min were best to improve the C/N and C/P in the secondary sludge. In the primary sludge, the ultrasonic power at 600 W, and the ultrasonic time at 15 min were best to improve C/N and C/P. This study lays a foundation for the practical application of ultrasonic pretreatment of sludge and provides basic information for typical industrial scenarios.

ML Approaches for the Study of Significant Heritage Contexts: An Application on Coastal Landscapes in Sardinia

Remote Sensing (RS) and Geographic Information Science (GIS) techniques are powerful tools for spatial data collection, analysis, management, and digitization within cultural heritage frameworks. Despite their capabilities, challenges remain in automating data semantic classification for conservation purposes. To address this, leveraging airborne Light Detection And Ranging (LiDAR) point clouds, complex spatial analyses, and automated data structuring is crucial for supporting heritage preservation and knowledge processes. In this context, the present contribution investigates the latest Artificial Intelligence (AI) technologies for automating existing LiDAR data structuring, focusing on the case study of Sardinia coastlines. Moreover, the study preliminary addresses automation challenges in the perspective of historical defensive landscapes mapping. Since historical defensive architectures and landscapes are characterized by several challenging complexities—including their association with dark periods in recent history and chronological stratification—their digitization and preservation are highly multidisciplinary issues. This research aims to improve data structuring automation in these large heritage contexts with a multiscale approach by applying Machine Learning (ML) techniques to low-scale 3D Airborne Laser Scanning (ALS) point clouds. The study thus develops a predictive Deep Learning Model (DLM) for the semantic segmentation of sparse point clouds (<10 pts/m2), adaptable to large landscape heritage contexts and heterogeneous data scales. Additionally, a preliminary investigation into object-detection methods has been conducted to map specific fortification artifacts efficiently.

Stability in Sight: Leveraging Machine Learning for Proactive Political Risk Management in the United States of America

This paper explores the application of machine learning (ML) in political risk management, with a specific focus on recent trends in political violence in the United States OF America. The growing intersection of political polarization, disinformation, and societal unrest has created a volatile political climate, as evidenced by events such as the January 6 Capitol insurrection and rising threats to public officials. The paper argues that machine learning could play a critical role in mitigating such risks by analyzing large datasets, including social media interactions, political speeches, and public sentiment, to predict potential flashpoints of violence. Through predictive analytics, sentiment analysis, and anomaly detection, ML can enhance decision-making processes and provide timely interventions to avert violent incidents. Additionally, case studies demonstrate ML’s superiority over traditional methods in risk assessments. Despite the challenges associated with ML, such as data privacy concerns, algorithmic bias, and the complexity of political contexts, this paper argues that machine learning holds immense potential in transforming political risk management. By integrating diverse data sources and refining risk models, ML can significantly improve accuracy and efficiency in predicting and mitigating political risks. The paper concludes with recommendations for further integrating ML tools in political risk strategies to address the increasingly unstable political environment.

Performance of Predictive Models in Cervical Cancer Recurrence: A Systematic Review and Meta-Analysis of Biomarkers and Prognosis

This study investigates the use of Predictive Analytics Frameworks (PAF) for identifying biomarkers of recurrent cervical cancer and predicting prognosis. It addresses the limited comprehensive evaluations of the effectiveness of predictive models in this area, despite the growing application of machine learning in healthcare. The purpose of this systematic review and meta-analysis is to assess the performance of predictive analytics models in terms of sensitivity, accuracy, specificity, and the area under the curve (AUC-ROC) for identifying cervical cancer biomarkers and predicting prognosis. To address this research problem, a systematic review and meta-analysis was conducted, covering studies published between 2014 and 2024. A total of 1,515 studies were initially identified from the PubMed and Scopus databases, with 50 research studies meeting the inclusion criteria. Repeated measures ANOVA and meta-analysis were applied using data collected over an 8-year period to evaluate recurrence trends and the predictive power of various models. The findings suggest that predictive analytics models show significant potential for improving diagnostic accuracy in identifying cervical cancer biomarkers. However, the review also highlights several limitations, including the small number of included studies, heterogeneity across studies, and potential bias in retrospective analyses. In conclusion, while predictive analytics frameworks demonstrate promise in improving cervical cancer prognosis and biomarker identification, further research is required to validate these findings and assess their broader clinical utility. The study underscores the importance of continued exploration of predictive models to enhance decision-making in oncology.

Practical application of machine learning in energy and thermal management: Long-term data analysis of solar-assisted AC systems in portable cabins in Kuwait and Australia

Machine learning techniques are advancing rapidly in theoretical grounds, but their practical application and real evaluation in engineering disciplines are limited. This study investigates one-year of experimental data for indoor air conditions and energy monitoring of two identical portable cabins in Kuwait. Additionally, a 9-month period of solar photovoltaic energy production is experimentally obtained for assessing energy saving aspects of a solar assisted air conditioning system in one of the cabins. A transient system simulation tool model is developed and validated for the portable cabins. Both one-year experimental data and the validated simulation data are used for developing machine learning models for six case studies in Kuwait and Australia. In simulations, there are issues related to lack of access to real-time weather data from a locally installed weather station. In experimental works, there were issues related to disruption of data collection due to power or internet shutdowns. To resolve these issues, nineteen regression models are investigated. It was found that all models performed very well although the Matern 5/2 and Exponential gaussian process regression model performed slightly better, with all models achieving high accuracy, as indicated by R-squared values close to 1.0 in all six case studies. Overall, it was found that the photovoltaic solar-assisted air conditioning system could save annually 26.2 % and 75.7 % energy in Kuwait and Rockhampton, Australia, respectively.

Comparative Study on Stock Movement Prediction Using Hybrid Deep Learning Model

Applying machine learning techniques in stock market prediction has evolved significantly, with deep learning methodologies gaining prominence. Conventional algorithms such as Linear Regression and Neural Networks initially dominated but struggled to capture complex temporal dependencies in financial data. Recent research has explored deep learning architectures like LSTM and CNN and hybrids such as CNN-LSTM and LSTM-CNN, showcasing promising results. However, there's a gap in research comparing these models across different datasets, particularly in predicting stock movements. This study addresses this gap by conducting a comparative analysis of deep learning and hybrid models for stock movement prediction in the Indonesian banking sector. The evaluation based on RMSE and MAE reveals that the LSTM-CNN hybrid consistently outperforms other models, showcasing its versatility and accuracy across different data characteristics. Then, exploration through hyperparameter tuning demonstrates the criticality of parameter selection in optimizing model performance. These findings contribute to advancing predictive modeling in financial markets, offering valuable insights for investors, analysts, and policymakers. Further research in hyperparameter tuning and model optimization holds promise for enhancing accuracy and reliability in stock price prediction.

Efficient machine learning models across multiple datasets for autism spectrum disorder diagnoses

Autism Spectrum Disorder (ASD) presents a complex neurodevelopmental challenge with diverse behavioral symptoms that overlap with other mental disorders, complicating diagnosis. Challenges in existing systems include the complexity and diversity of ASD symptoms, as well as the need for innovative and efficient detection strategies across different age groups. The application of machine learning (ML) offers a promising avenue for revolutionizing ASD prediction and diagnosis. The proposed work aims to overcome these challenges by leveraging ML and Automated Machine Learning (AutoML) techniques, combined with comprehensive datasets and advanced analytics, to enhance ASD detection and prediction capabilities. This study introduces the Autism Spectrum Disorder Detection (ASDD) Framework, which systematically applies ML and AutoML across text-based datasets for various age groups. A combined dataset is generated through the concatenation of various data sources, enriching the diagnostic scope. The framework incorporates AutoML tools such as Lazy Predict, TPOT, and AutoKeras, alongside feature selection methods like Principal Component Analysis (PCA) and Chi-Square. It also employs robust hyperparameter optimization techniques, including Bayesian optimization, Random Search, and genetic algorithms, and uses advanced ensemble methods like voting and bagging to boost prediction accuracy. An empirical evaluation of 27 distinct ML models across seven datasets, including a synthesized dataset named DB8, demonstrates exceptional performance, with many models achieving 100% accuracy and perfect F1 scores in various contexts. These outcomes highlight the effectiveness of the proposed framework in advancing ASD detection and prediction methodologies.

Using dynamic time warping to measure similarity between spectra

Many situations in acoustics and noise control involve comparing two or more spectra. For example, when tuning a numerical model to match measured data, the measured and predicted sound pressure spectra will be compared. In machine learning applications, it is useful to express the similarity between two spectra as a single number, which can then be incorporated into a cost function. It is common to use the mean square difference between the two spectra as this number. However, the mean square difference does not always reflect an acoustician's intuition of the similarity between two spectra. For example, if the natural frequencies of the two spectra are slightly misaligned, the mean square difference can be quite high, even though the underlying system parameters are similar. In this paper, several different similarity metrics for spectra are explored. Most notably, dynamic time warping (DTW) and its variants are applied to spectra and compared to more traditional metrics. Modifications to the DTW algorithm to handle logarithmic frequency and amplitude spacing as commonly arise when considering spectra are also addressed.

Machine learning in oncological pharmacogenomics: advancing personalized chemotherapy.

This review analyzes the application of machine learning (ML) in oncological pharmacogenomics, focusing on customizing chemotherapy treatments. It explores how ML can analyze extensive genomic, proteomic, and other omics datasets to identify genetic patterns associated with drug responses. This, in turn, facilitates personalized therapies that are more effective and have fewer side effects. Recent studies have emphasized ML's revolutionary role of ML in personalized oncology treatment by identifying genetic variability and understanding cancer pharmacodynamics. Integrating ML with electronic health records and clinical data shows promise in refining chemotherapy recommendations by considering the complex influencing factors. Although standard chemotherapy depends on population-based doses and treatment regimens, customized techniques use genetic information to tailor treatments for specific patients, potentially enhancing efficacy and reducing adverse effects.However, challenges, such as model interpretability, data quality, transparency, ethical issues related to data privacy, and health disparities, remain. Machine learning has been used to transform oncological pharmacogenomics by enabling personalized chemotherapy treatments. This review highlights ML's potential of ML to enhance treatment effectiveness and minimize side effects through detailed genetic analysis. It also addresses ongoing challenges including improved model interpretability, data quality, and ethical considerations. The review concludes by emphasizing the importance of rigorous clinical trials and interdisciplinary collaboration in the ethical implementation of ML-driven personalized medicine, paving the way for improved outcomes in cancer patients and marking a new frontier in cancer treatment.

CDT1 is a Potential Therapeutic Target for the Progression of NAFLD to HCC and the Exacerbation of Cancer

Aims: This study aimed to identify potential therapeutic targets in the progression from non-alcoholic fatty liver disease (NAFLD) to hepatocellular carcinoma (HCC), with a focus on genes that could influence disease development and progression. Background: Hepatocellular carcinoma, significantly driven by non-alcoholic fatty liver disease, represents a major global health challenge due to late-stage diagnosis and limited treatment options. This study utilizes bioinformatics to analyze data from GEO and TCGA, aiming to uncover molecular biomarkers that bridge NAFLD to HCC. Through identifying critical genes and pathways, our research seeks to advance early detection and develop targeted therapies, potentially improving prognosis and personalizing treatment for NAFLD-HCC patients. Objective: Identify key genes that differ between NAFLD and HCC; Analyze these genes to understand their roles in disease progression; Validate the functions of these genes in NAFLD to HCC transition. Methods: Initially, we identified a set of genes differentially expressed in both NAFLD and HCC using second-generation sequencing data from the GEO and TCGA databases. We then employed a Cox proportional hazards model and a Lasso regression model, applying machine learning techniques to the large sample data from TCGA. This approach was used to screen for key disease-related genes, and an external dataset was utilized for model validation. Additionally, pseudo-temporal sequencing analysis of single-cell sequencing data was performed to further examine the variations in these genes in NAFLD and HCC. Results: The machine learning analysis identified IGSF3, CENPW, CDT1, and CDC6 as key genes. Furthermore, constructing a machine learning model for CDT1 revealed it to be the most critical gene, with model validation yielding an ROC value greater than 0.80. The single-cell sequencing data analysis confirmed significant variations in the four predicted key genes between the NAFLD and HCC groups. Conclusion: Our study underscores the pivotal role of CDT1 in the progression from NAFLD to HCC. This finding opens new avenues for early diagnosis and targeted therapy of HCC, highlighting CDT1 as a potential therapeutic target.

A unified active learning framework for annotating graph data for regression tasks

In many domains, effectively applying machine learning models requires a large number of annotations and labelled data, which might not be available in advance. Acquiring annotations often requires significant time, effort, and computational resources, making it challenging. Active learning strategies are pivotal in addressing these challenges, particularly for diverse data types such as graphs. Although active learning has been extensively explored for node-level classification, its application to graph-level learning, especially for regression tasks, is not well-explored. We develop a unified active learning framework specializing in graph annotating and graph-level learning for regression tasks on both standard and expanded graphs, which are more detailed representations. We begin with graph collection and construction. Then, we construct various graph embeddings (unsupervised and supervised) into a latent space. Given such an embedding, the framework becomes task agnostic and active learning can be performed using any regression method and query strategy suited for regression. Within this framework, we investigate the impact of using different levels of information for active and passive learning, e.g., partially available labels and unlabelled test data. Despite our framework being domain agnostic, we validate it on a real-world application of software performance prediction, where the execution time of the source code is predicted. Thus, the graph is constructed as an intermediate source code representation. We support our methodology with a real-world dataset to underscore the applicability of our approach. Our real-world experiments reveal that satisfactory performance can be achieved by querying labels for only a small subset of all the data. A key finding is that Graph2Vec (an unsupervised embedding approach for graph data) performs the best, but only when all train and test features are used. However, Graph Neural Networks (GNNs) are the most flexible embedding techniques when used for different levels of information with and without label access. In addition, we find that the benefit of active learning increases for larger datasets (more graphs) and when the graphs are more complex, which is arguably when active learning is the most important.

Enhancing dietary analysis: Using machine learning for food caloric and health risk assessment.

In the wake of growing concerns regarding diet-related health issues, this study investigates the application of machine learning methods to estimate the energy content and classify the health risks of foods based on the USDA National Nutrient Database. The caloric content of foods was estimated using the nutritional composition (i.e., carbohydrates, protein, total lipid, and total sugar content) and classified based on their weighted health risks, considering their carbohydrate, lipid, and glycemic index levels. The algorithms used for modeling include multiple linear regression (MLR), K-nearest neighbors, support vector machine, random forest regression (RFR), gradient-boosted regression, decision trees (DT), and deep neural networks. The MLR model demonstrated high accuracy on the training dataset (R2=0.99, mean absolute error [MAE]=7.71kcal, and root mean squared error [RMSE]=17.89kcal) and testing dataset (R2=0.99, MAE=7.75kcal, and RMSE=18kcal) in energy estimation, indicating its effectiveness in dietary assessment. The RFR and DT models were useful in categorizing foods into low-health-risk foods, but their performance was reduced in medium and high-health-risk groups. This research contributes to developing tools that could aid in personalized dietary planning and public health interventions to mitigate diet-related health risks. PRACTICAL APPLICATION: This study applies machine learning to estimate how many calories are in food and to understand the health risks different foods might have. By investigating the fats, cholesterol, and sugars in food items listed in a public database, we can better plan diets or develop apps that help people make healthier eating choices. This work aims to improve public access to nutritional information, supporting efforts to combat diet-related diseases through educational materials and applications that guide dietary choices in various settings.

Current scenario of machine learning applications to hydrothermal liquefaction via bibliometric analysis

Background Energy shortages and global warming have been significant issues throughout history. Therefore, the search for environmentally friendly renewable energy sources is crucial for achieving sustainability. Biomass energy is gaining global attention as a renewable energy option, particularly through the process of hydrothermal liquefaction, which converts biomass into bio-crude oil. Methods Hydrothermal liquefaction is a complex process that is challenging to explain, leading to research on machine learning models for this process. These models aim to predict values and investigate the impact of variables on the hydrothermal liquefaction process. However, the development of machine learning in hydrothermal liquefaction is still limited due to its novelty and the time required for comprehensive study. Thus, the objective of this study was to analyze relevant publications in the Scopus database, focusing on indexed ML and HTL keywords, to understand keyword associations and co-citations. Results The results reveal an increasing trend in the study of ML in the HTL process, with a growing interest from various countries. Conclusion Notably, China currently holds the largest share of ML research in HTL processes, with most published works falling within the field of engineering. The keyword “liquefaction” emerges as the most popular term in these publications.

Evaluating the impact of industrial wastes on the compressive strength of concrete using closed-form machine learning algorithms

Industrial wastes have found great use in the built environment due to the role they play in the sustainable infrastructure development especially in green concrete production. In this research investigation, the impact of wastes from the industry on the compressive strength of concrete incorporating fly ash (FA) and silica fume (SF) as additional components alongside traditional concrete mixes has been studied through the application of machine learning (ML). A green concrete database comprising 330 concrete mix data points has been collected and modelled to estimate the unconfined compressive strength behaviour. Considering the concerning environmental ramifications associated with concrete production and its utilization in construction activities, there is a pressing need to perform predictive model exercise. Furthermore, given the prevalent reliance of concrete production professionals on laboratory experiments, it is imperative to propose smart equations aimed at diminishing this dependency. These equations should be applicable for use in the design, construction, and performance assessment of concrete infrastructure, thereby reflecting the multi-objective nature of this research endeavour. It has been proposed by previous research works that the addition of FA and SF in concrete has a reduction impact on the environmental influence indicators due to reduced cement use. The artificial neural network (ANN) and the M5P models were applied in this exercise to predict the compressive strength of FA- and SF-mixed concrete also considering the impact of water reducing agent in the concrete. A sensitivity analysis was also conducted to determine the impact of the concrete components on the strength of the concrete. At the end, closed-form equations were proposed by the ANN and M5P with performance indices which outperformed previous models conducted on the same database size. The result of the sensitivity analysis showed that FA is most impactful of all the studied components thereby emphasizing the importance of adding industrial wastes in concrete production for improved mechanical properties and reduced carbon footprint in the concrete construction activities. Also, the M5P and ANN models with R2 of 0.99 showed a potential for use as decisive models to predict the compressive strength of FA- and SF-mixed concrete.

A COMPREHENSIVE REVIEW OF MACHINE LEARNING AND DEEP LEARNING APPLICATIONS IN CYBERSECURITY: AN INTERDISCIPLINARY APPROACH

Cybersecurity is increasingly becoming a critical concern as the complexity and frequency of cyber-attacks continue to rise. Machine learning (ML) and deep learning (DL) have emerged as powerful tools to enhance cybersecurity systems, offering dynamic capabilities in real-time threat detection, anomaly detection, and intrusion prevention. This article (45) presents a systematic review of the applications of ML and DL in cybersecurity, adhering to the PRISMA guidelines. The review covers several key domains, including network security, cloud security, and Internet of Things (IoT) security, highlighting how ML/DL models like convolutional neural networks (CNNs) and recurrent neural networks (RNNs) outperform traditional rule-based systems. It also addresses challenges such as adversarial attacks, data privacy concerns, and the computational resource demands of DL models. Current solutions like adversarial training, federated learning, and model optimization techniques are examined for their potential to mitigate these issues. The findings suggest that while ML/DL technologies hold great promise, further research and innovation are necessary to overcome the inherent challenges, ensuring that these systems can be deployed effectively and securely in real-world environments.

Transforming Crop Management Through Advanced AI and Machine Learning: Insights into Innovative Strategies for Sustainable Agriculture

The integration of artificial intelligence (AI) and machine learning (ML) into crop management is transforming modern agriculture by enhancing efficiency, sustainability, and resilience. This review explores the multifaceted applications of AI and ML in key areas such as precision farming, pest and disease management, and harvest optimization. The use of AI-driven predictive analytics allows for more accurate forecasting of crop yields, pest outbreaks, and weather conditions, enabling farmers to make data-driven decisions that optimize resource use and reduce environmental impacts. A significant advancement is the integration of AI and ML with the Internet of Things (IoT) and autonomous farming equipment. These technologies enable real-time monitoring and precise interventions, enhancing productivity and minimizing labor costs. In crop breeding and genomics, AI accelerates the development of resilient crop varieties, which is crucial for adapting to climate change and increasing food demands. Despite the many benefits, challenges such as data quality, infrastructure limitations, and high implementation costs remain. The adoption of AI technologies is uneven, with small-scale farmers in developing regions facing barriers due to limited access to data and resources. Ethical concerns related to data privacy and the digital divide must also be addressed to ensure equitable access to these technologies. The future of AI and ML in agriculture lies in the development of more advanced predictive models, enhanced integration with the IoT, and the widespread use of autonomous farming systems.

Trends of machine learning for dental caries research in Southeast Asia: insights from a bibliometric analysis

Background Dental caries is a common chronic oral disease, posing a serious public health issue. By analyzing large datasets, machine learning shows potential in addressing this problem. This study employs bibliometric analysis to explore emerging topics, collaborations, key authors, and research trends in Southeast Asia related to the application of machine learning in dental caries management. Methods A comprehensive selection using the Scopus database to obtain relevant research, covering publications from inception to July 2024 was done. We employed the Bibliometric approaches, including co-authorship networks, yearly publishing trends, institutional and national partnerships, keyword co-occurrence analysis, and citation analysis, for the collected data. To explore the visualization and network analysis, we employed the tools such as VOSviewer and Bibliometrix in R package. Results The final bibliometric analysis included 246 papers. We found that Malaysia became the top contributor with 59 publications, followed by Indonesia (37) and Thailand (29). Malaysia had the highest Multiple Country Publications (MCP) ratio at 0.407. Top institutions including the Universiti Sains Malaysia led with 39 articles, followed by Chiang Mai University (36) and the National University of Singapore (30) became the leader. Co-authorship analysis using VOSviewer revealed six distinct clusters. A total of 1220 scholars contributed to these publications. The top 10 keywords, including ‘human’ and ‘dental caries,’ indicated research hotspots. Conclusion We found growing evidence of machine learning applications to address dental caries in Southeast Asia. The bibliometric analysis highlights key authors, collaborative networks, and emerging topics, revealing research trends since 2014. This study underscores the importance of bibliometric analysis in tackling this public health issue.