Data-driven Approach Research Articles

Machine learning­based approaches for enhancing human resource management using automated employee performance prediction systems

Purpose This study focuses on enhancing the accuracy and efficiency of employee performance prediction to enhance decision making and improve organisational productivity. By introducing advance machine learning (ML) techniques, this study aims to create a more reliable and data-driven approach to evaluate employee performance. Design/methodology/approach In this study, nine machine learning (ML) models were used for forecasting employee performance: Random Forest, AdaBoost, CatBoost, LGB Classifier, SVM, KNN, XGBoost, Decision Tree and one Hybrid model (SVM + XGBoost). Each ML model is trained on an HR data set covering various features such as employee demographics, job-related factors and past performance records, ensuring reliable performance predictions. Feature scaling techniques, namely, min-max scaling, Standard Scaler and PCA, have been used to enhance the effectiveness of employee performance prediction. The models are trained to classify data, predicting whether an employee’s performance meets expectations or needs improvement. Findings All proposed models used in the study can correctly categorize data with an average accuracy of 94%. Notably, the Random Forest model demonstrates the highest accuracy across all three scaling techniques, achieving optimise accuracy, respectively. The results presented have significant implications for HR procedures, providing businesses with the opportunity to make data-driven decisions, improve personnel management and foster a more effective and productive workforce. Research limitations/implications The scope of the used data set limits the study, despite our models delivering high accuracy. Further research could extend to different data sets or more diverse organisational settings to validate the model’s effectiveness across various contexts. Practical implications The proposed ML models in the study provide essential tools for HR departments, enabling them to make more informed data driven decisions with regard to employee performance. This approach can enhance personnel management, improve workforce productivity and fostering a more effective organisational environment. Social implications Although AI models have shown promising outcomes, it is crucial to recognise the constraints and difficulties involved in their use. To ensure the fair and responsible use of AI in employee performance prediction, ethical considerations, privacy problems and any biases in the data should be properly addressed. Future work will be required to improve and broaden the capabilities of AI models in predicting employee performance. Originality/value This study introduces an exclusive combination of ML models for accurately predicting employee performance. By employing these advanced techniques, the study offers novel insight into how organisations might transition from a conventional evaluation method to a more advanced and objective, data-backed approach.

Efficient Workflow Scheduling for Minimizing Data Transfers and Enhancing Resource Utilization in Cloud IaaS Platforms

Cloud IaaS platforms readily provide access to homogeneous multi-core machines, whether they are physical ("bare metal") or virtual machines. Each of these machines can be equipped with high-performance SSD disks, enabling the distribution of workflow-generated files across multiple machines, which helps minimize the overhead associated with data transfers. In this paper, we propose a scheduling algorithm called SMDT-ERU (Scheduling for Minimizing Data Transfer - Enhancing Resource Utilization), designed to reduce the makespan of data-intensive workflows by minimizing data transfers between dependent tasks over the network. Intermediate files generated by tasks are stored locally on the disk of the machine where the tasks are executed. Through experimentation, we confirm that increasing the number of cores per machine reduces the additional costs caused by network data transfers. Real-world workflow experiments demonstrate the advantages of the proposed algorithm. Our data-driven scheduling approach significantly reduces execution time and the volume of data transferred over the network, outperforming one of the leading state-of-the-art algorithms, which we have adapted to fit our assumptions.

Conceptual advances in petrophysical inversion techniques: The synergy of machine learning and traditional inversion models

Petrophysical inversion techniques are essential for estimating reservoir properties such as porosity, permeability, and fluid saturation, which are critical for optimizing hydrocarbon recovery and reservoir management. Traditionally, deterministic and stochastic inversion methods have relied on well log data, core samples, and seismic information to generate models of subsurface properties. However, these techniques face limitations in handling large datasets, nonlinear relationships, and uncertainties in complex reservoir environments. Recent advances in machine learning (ML) offer an opportunity to enhance these traditional methods by providing a data-driven approach capable of learning patterns and refining inversion models. This review explores the potential synergy between traditional petrophysical inversion techniques and machine learning algorithms to create a new paradigm for reservoir parameter estimation and analysis. Machine learning’s ability to process large volumes of diverse data and capture nonlinear relationships complements the physics-based constraints of traditional inversion methods. By combining these two approaches, hybrid models can provide improved accuracy in predicting reservoir properties and handling data gaps or uncertainties. The review discusses specific case studies where machine learning-assisted inversion models have been successfully applied, such as the estimation of porosity and permeability, seismic inversion for lithology identification, and fluid saturation modeling. These hybrid models demonstrate enhanced predictive accuracy, faster computational efficiency, and real-time adaptability, transforming reservoir characterization and management. The integration of machine learning with traditional inversion techniques represents a major advancement in petrophysical analysis. This synergy holds the potential to revolutionize reservoir parameter estimation, making it more accurate, robust, and responsive to real-time data, ultimately improving decision-making in reservoir engineering and maximizing hydrocarbon recovery. Keywords: Petrophysical Inversion, Machine Learning, Traditional Inversion, Models.

Knowledge Aware Automated Health Claims Processing with Medical Ontologies and Large Language Models

One of the major challenges in automating health insurance claims processing lies in the complexity involved in validating an incoming claim's medical diagnoses against its policy Underwriting (UW) exclusions. Termed UW Exclusion Detection, this process ensures claims are only paid out if their diagnoses are not medically associated with conditions excluded under the policy. Medical diagnoses in health insurance claims are typically represented by the International Classification of Disease (ICD) codes, established by the World Health Organization. For example, given a policy that excludes "all respiratory illness". A claim with the ICD code J45 (Asthma) will be subject to rejection as J45 is a respiratory-related diagnosis that falls within the scope of the policy's UW exclusion. The key challenge in automating this process lies in the wide range of available ICD codes. The ICD-10-CM coding scheme consists of over 40,000 codes, which often results in scenarios where codes encountered during inference are absent from the training data. These unseen ICD codes limit the effectiveness of data-driven approaches, which depend on the training data to discern medically relevant associations between UW exclusions and ICD codes. This underscores the need to supplement data-driven approaches with additional domain knowledge. We hypothesize that integrating implicit medical domain knowledge inherent in Large Language Models (LLMs) with explicit domain knowledge from medical ontologies, will enhance data-driven approaches for UW Exclusion Detection. Thoroughly validated on real-world health insurance claims data, our proposed approach proved effective in accurately establishing medically relevant associations between UW exclusions and ICD codes.

Optimizing Agile, Quality-assured Development of AD Software through Data-driven Approaches

Optimizing Agile, Quality-assured Development of AD Software through Data-driven Approaches

The Role of Artificial Intelligence in Revolutionizing Mental Health Services: A Data-Driven Approach

The potential of artificial intelligence (AI) to transform mental health care through individualized treatment approaches is significant. This research investigates the application of AI and extensive healthcare datasets to improve the precision of mental health diagnoses and treatments. The study employs various machine learning techniques, such as Random Forest, CatBoost, K-Nearest Neighbors, XGBoost, Convolutional Neural Networks, and Long Short-Term Memory networks, to analyze diverse data sources including electronic health records, neuroimaging, genetic information, and demographic data. Following thorough data preprocessing, model training, and evaluation using metrics like accuracy, precision, recall, F1 score, and Cohen's Kappa, the Random Forest algorithm emerges as the top performer with 98.7% accuracy. The research highlights the importance of addressing ethical considerations, such as protecting patient privacy, ensuring data security, and mitigating algorithmic bias when implementing AI in mental health services. The findings indicate that AI-based approaches can markedly improve diagnostic accuracy, provide tailored treatment suggestions, and facilitate early relapse detection, thus promoting proactive mental health management. Key Words: Artificial Intelligence, Mental Health, Data- Driven approach, Personalized Treatment, Machine Learning, Natural Language Processing.

Machine Learning-Assisted Hartree–Fock Approach for Energy Level Calculations in the Neutral Ytterbium Atom

Data-driven machine learning approaches with precise predictive capabilities are proposed to address the long-standing challenges in the calculation of complex many-electron atomic systems, including high computational costs and limited accuracy. In this work, we develop a general workflow for machine learning-assisted atomic structure calculations based on the Cowan code’s Hartree–Fock with relativistic corrections (HFR) theory. The workflow incorporates enhanced ElasticNet and XGBoost algorithms, refined using entropy weight methodology to optimize performance. This semi-empirical framework is applied to calculate and analyze the excited state energy levels of the 4f closed-shell Yb I atom, providing insights into the applicability of different algorithms under various conditions. The reliability and advantages of this innovative approach are demonstrated through comprehensive comparisons with ab initio calculations, experimental data, and other theoretical results.

Combating Antimicrobial Resistance Through a Data-Driven Approach to Optimize Antibiotic Use and Improve Patient Outcomes: Protocol for a Mixed Methods Study

Background It is projected that drug-resistant infections will lead to 10 million deaths annually by 2050 if left unabated. Despite this threat, surveillance data from resource-limited settings are scarce and often lack antimicrobial resistance (AMR)–related clinical outcomes and economic burden. We aim to build an AMR and antimicrobial use (AMU) data warehouse, describe the trends of resistance and antibiotic use, determine the economic burden of AMR in Uganda, and develop a machine learning algorithm to predict AMR-related clinical outcomes. Objective The overall objective of the study is to use data-driven approaches to optimize antibiotic use and combat antimicrobial-resistant infections in Uganda. We aim to (1) build a dynamic AMR and antimicrobial use and consumption (AMUC) data warehouse to support research in AMR and AMUC to inform AMR-related interventions and public health policy, (2) evaluate the trends in AMR and antibiotic use based on annual antibiotic and point prevalence survey data collected at 9 regional referral hospitals over a 5-year period, (3) develop a machine learning model to predict the clinical outcomes of patients with bacterial infectious syndromes due to drug-resistant pathogens, and (4) estimate the annual economic burden of AMR in Uganda using the cost-of-illness approach. Methods We will conduct a study involving data curation, machine learning–based modeling, and cost-of-illness analysis using AMR and AMU data abstracted from procurement, human resources, and clinical records of patients with bacterial infectious syndromes at 9 regional referral hospitals in Uganda collected between 2018 and 2026. We will use data curation procedures, FLAIR (Findable, Linkable, Accessible, Interactable and Repeatable) principles, and role-based access control to build a robust and dynamic AMR and AMU data warehouse. We will also apply machine learning algorithms to model AMR-related clinical outcomes, advanced statistical analysis to study AMR and AMU trends, and cost-of-illness analysis to determine the AMR-related economic burden. Results The study received funding from the Wellcome Trust through the Centers for Antimicrobial Optimisation Network (CAMO-Net) in April 2023. As of October 28, 2024, we completed data warehouse development, which is now under testing; completed data curation of the historical Fleming Fund surveillance data (2020-2023); and collected retrospective AMR records for 599 patients that contained clinical outcomes and cost-of-illness economic burden data across 9 surveillance sites for objectives 3 and 4, respectively. Conclusions The data warehouse will promote access to rich and interlinked AMR and AMU data sets to answer AMR program and research questions using a wide evidence base. The AMR-related clinical outcomes model and cost data will facilitate improvement in the clinical management of AMR patients and guide resource allocation to support AMR surveillance and interventions. International Registered Report Identifier (IRRID) PRR1-10.2196/58116

Data and model synergy-driven rolling bearings remaining useful life prediction approach based on deep neural network and Wiener process

Abstract Various Remaining Useful Life (RUL) prediction methods, encompassing model-based, data-driven, and hybrid methods, have been developed and successfully applied to prognostics and health management for diverse rolling bearing. Hybrid methods that integrate the advantages of model-based and data-driven approaches have garnered significant attention. However, the effective integration of the two methods to address the randomness in rolling bearing full lifecycle processes remains a significant challenge. To overcome the challenge, this paper proposes a data and model synergy-driven RUL prediction framework that includes two data and model synergy approaches. Firstly, a convolutional stacked bidirectional long and short-term memory network with temporal attention mechanism is established to construct health index. The RUL prediction is achieved based on HI and polynomial model. Secondly, a three-phase prediction model based on the Wiener process is established by considering the evolution law of different degradation phases. Then, two synergy methods are designed. Strategy 1: HI is adopted as the observation value for online updating of physics degradation model parameters under Bayesian framework, and the RUL prediction results are obtained from the physics degradation model. Strategy 2: The RUL prediction results from the data-driven and physics-based model are weighted linearly combined to improve the overall prediction accuracy. The effectiveness of the proposed model is verified using two bearing full lifecycle datasets. The results indicate that the proposed approach can accommodate both short-term and long-term RUL predictions, outperforming state-of-the-art single models.

Optimal dispatch scheme considering system operational flexibility

To enhance the system's capacity to manage emergencies and foster the efficient utilization of newly generated energy, this paper introduces a redefined concept of system operation flexibility and proposes a two-stage optimized economic dispatch model. This model is tailored to ensure the system's operational flexibility. In this method, net load is pinpointed as the primary driver of flexibility demand, and a data-driven approach is employed to capture the uncertainties associated with net load across various risk levels. As flexibility supplier, conventional units work in tandem to optimize their operating points, fulfilling flexibility demand while adhering to the Do Not Exceed (DNE) baseline, thereby ensuring that system flexibility remains within safe limits. A case study on the modified SG-126 bus system demonstrates that the proposed methodology effectively highlights flexibility shortfalls and validates its efficiency in terms of cost and inter-regional coordination.

Machine learning insights into regional dynamics and prevalence of COVID-19 variants in US health and human services regions

BackgroundThe COVID-19 pandemic arising from the emergence of SARS-CoV-2 in late 2019 has led to global devastation with millions of lives lost by January 2024. Despite the WHO’s declaration of the end of the global health emergency in May 2023, the virus persists, propelled by mutations. Variants continue to challenge vaccination efforts, underscoring the necessity for ongoing vigilance. This study aimed at contributing to a more data-driven approach to pandemic management by employing random forest regression to analyze regional variant prevalence.MethodsThis study utilized data from various sources including National COVID Cohort Collaborative database, Bureau of Transportation Statistics, World Weather Online, EPA, and US Census. Key variables include pollution, weather, travel patterns, and demographics. Preprocessing steps involved merging and normalization of datasets. Training data spanned from January 2021 to February 2023. The Random Forest (RF) Regressor was chosen for its accuracy in modeling. To prevent data leakage, time series splits were employed. Model performance was evaluated using metrics such as MSE and R-squared.ResultsThe Alpha variant was predominant in the Southeast, with less than 80% share even at its peak. Delta surged initially in Kansas City and maintained dominance there for over 5 months. Omicron subvariant BA.5 spread nationwide, becoming predominant across all Health and Human Services regions simultaneously, with New York seeing the earliest and fastest decline in its share. Variant XBB.1.5 concentrated more in the Northeast, but limited data hindered full analysis. Using RF Regressor, key features affecting spread patterns were identified, with high predictive accuracy. Each variant showed specific environmental correlations; for instance, Alpha with air quality index, Delta with ozone density, BA.5 with UV index, and XBB.1.5 with land area and income. Correlation analysis further highlighted variant-specific associations.ConclusionsThis research provides a comprehensive analysis of the regional distribution of COVID-19 variants, offering critical insights for devising targeted public health strategies. By utilizing machine learning, the study uncovers the complex factors contributing to variant spread and reveals how specific factors contribute to variant prevalence, offering insights crucial for pandemic management.

Elite Politics, Mass Discontent and Political Inequality in South Korea: Who Represents Me?

In an era of heightened public skepticism toward political institutions, understanding the perceived divide between political elites and the general population has become increasingly essential. This study examines how Korean citizens perceive their political representatives, focusing on the socio-economic and demographic disparities between political elites and the wider public. To achieve this, this study concentrates on three core elements: citizens’ perceptions of politicians, the social networks of elite politicians, and the socio-economic and demographic attributes of these elites in comparison to the general population. Employing a data-driven approach—including text analysis, clustering, social network analysis, and visualization techniques—this study investigates data on political elites alongside district-level data. The findings reveal significant socio-economic and demographic disparities between political elites and the public, with elites predominantly emerging from narrow backgrounds characterized by affluence, high education, and an overrepresentation of males above the age of 60. These results underscore the need for measures that promote transparency, accountability, and inclusivity within the political system, highlighting the importance of bridging the divide between political elites and the general public to cultivate a more representative and responsive political environment.

Critical Assessment of Protein Engineering (CAPE): A Student Challenge on the Cloud.

The success of AlphaFold in protein structure prediction highlights the power of data-driven approaches in scientific research. However, developing machine learning models to design and engineer proteins with desirable functions is hampered by limited access to high-quality data sets and experimental feedback. The Critical Assessment of Protein Engineering (CAPE) challenge addresses these issues through a student-focused competition, utilizing cloud computing and biofoundries to lower barriers to entry. CAPE serves as an open platform for community learning, where mutant data sets and design algorithms from past contestants help improve overall performance in subsequent rounds. Through two competition rounds, student participants collectively designed >1500 new mutant sequences, with the best-performing variants exhibiting catalytic activity up to 5-fold higher than the wild-type parent. We envision CAPE as a collaborative platform to engage young researchers and promote computational protein engineering.

Data-Driven Approaches in Antimicrobial Resistance: Machine Learning Solutions

Background/Objectives: The emergence of antimicrobial resistance (AMR) due to the misuse and overuse of antibiotics has become a critical threat to global public health. There is a dire need to forecast AMR to understand the underlying mechanisms of resistance for the development of effective interventions. This paper explores the capability of machine learning (ML) methods, particularly unsupervised learning methods, to enhance the understanding and prediction of AMR. It aims to determine the patterns from AMR gene data that are clinically relevant and, in public health, capable of informing strategies. Methods: We analyzed AMR gene data in the PanRes dataset by applying unsupervised learning techniques, namely K-means clustering and Principal Component Analysis (PCA). These techniques were applied to identify clusters based on gene length and distribution according to resistance class, offering insights into the resistance genes’ structural and functional properties. Data preprocessing, such as filtering and normalization, was conducted prior to applying machine learning methods to ensure consistency and accuracy. Our methodology included the preprocessing of data and reduction of dimensionality to ensure that our models were both accurate and interpretable. Results: The unsupervised learning models highlighted distinct clusters of AMR genes, with significant patterns in gene length, including their associated resistance classes. Further dimensionality reduction by PCA allows for clearer visualizations of relationships among gene groupings. These patterns provide novel insights into the potential mechanisms of resistance, particularly the role of gene length in different resistance pathways. Conclusions: This study demonstrates the potential of ML, specifically unsupervised approaches, to enhance the understanding of AMR. The identified patterns in resistance genes could support clinical decision-making and inform public health interventions. However, challenges remain, particularly in integrating genomic data and ensuring model interpretability. Further research is needed to advance ML applications in AMR prediction and management.

Pioneering Museum Data Transformation: Unifying Legacy Systems and Enhancing Data Integrity

In the digital era, museums confront the challenge of modernising legacy data systems to align with current standards. Part of the RECODE (Rethinking Collections Data Ecosystems) Programme (Dupont et al. 2022) examines the complex transition from disparate departmental object models to a unified system, reflecting the broader museum community's struggle with data standardisation. The case study centres on consolidating five distinct data models accumulated over two decades, each tailored to different departmental needs under various constraints. The data mapping process identifies and aligns similar fields across models, facilitating the integration of analogous data points and exposing redundancies and unique practices embedded over years. Major challenges include data complexity, diversity, and quality issues (cleaning, standardizing, and deduplicating), compounded by the massive scale of the data: currently, the Natural History Museum London (NHM) has 262,831,590 records across 8,175 fields, and a total volume of 12 TB of data, which define both the ongoing data modelling work and our future data migration. A key innovation in our process is the immediate visibility of data issues that became apparent upon migrating to Amazon Web Services (AWS). AWS serves as the staging environment for our transition to the new NHM Collections Management System (CMS) and offers an unprecedented platform for directly addressing these challenges. It enables us to query all our data—across all departments and fields—in just seconds. This capability provides a comprehensive view that was previously unattainable. The core of this presentation is sharing "lessons learned" from navigating the intricacies of an ongoing CMS transition within a museum. The endeavor to untangle and unify diverse data models is a common challenge (IEEE Big Data Governance and Metadata Management Industry Connections Activity 2020,Wu et al. 2022, Wu et al. 2021,Little et al. 2022,Woodburn et al. 2022) highlighting the importance of community engagement and knowledge exchange. Our findings underscore the necessity of cross-departmental collaboration and the benefits of a data-driven data modelling approach. By sharing our journey and the developed comprehensive object model (Collier and Woodburn 2022), we aim to contribute valuable insights to museums undergoing similar transformations. The RECODE Programme sheds light on the practical aspects of CMS modernisation through the analytics and knowledge derived from over two decades of collections data and data use.

Predicting Feynman periods in ϕ4-theory

We present efficient data-driven approaches to predict the value of subdivergence-free Feynman integrals (Feynman periods) in ϕ4-theory from properties of the underlying Feynman graphs, based on a statistical examination of almost 2 million graphs. We find that the numbers of cuts and cycles determines the period to better than 2% relative accuracy. Hepp bound and Martin invariant allow for even more accurate predictions. In most cases, the period is a multi-linear function of the properties in question. Furthermore, we investigate the usefulness of machine-learning algorithms to predict the period. When sufficiently many properties of the graph are used, the period can be predicted with better than 0.05% relative accuracy.We use one of the constructed prediction models for weighted Monte-Carlo sampling of Feynman graphs, and compute the primitive contribution to the beta function of ϕ4-theory at L ∈ {13, … , 17} loops. Our results confirm the previously known numerical estimates of the primitive beta function and improve their accuracy. Compared to uniform random sampling of graphs, our new algorithm is 1000-times faster to reach a desired accuracy, or reaches 32-fold higher accuracy in fixed runtime.The dataset of all periods computed for this work, combined with a previous dataset, is made publicly available. Besides the physical application, it could serve as a benchmark for graph-based machine learning algorithms.

Riemannian Manifold-Aided Data-driven Diagnosis of Photovoltaic Hot Spots

Abstract This paper proposes a novel data-driven approach for hot-spot fault detection in photovoltaic (PV) modules, utilizing a curved Riemannian manifold (RM) to characterize the sample space. The proposed method detects hot spots in PV systems by mining geometric features in high-dimensional data. The RM-based method combines the geometric properties of Riemannian manifolds for fault detection. In addition, it also provides information about the severity of the hot spots. The proposed method has three main advantages:
1) it is capable of actively learning fault characteristics and can be applied to diagnose various degrees of hot spots;
2) unlike other data-driven methods, the proposed method considers non-Euclidean spatial data features, which further improves the accuracy of the method and reduces the false alarm rate (FAR);
3) it does not rely on mathematical models or expert knowledge, making it able to meet the real-time monitoring needs of actual PV systems.
The effectiveness and superiority of the proposed approach are verified through 12 sets of hot spots tests conducted on a PV experimental platform.

Assessing Helicopter Turbine Engine Health: A Simple Yet Robust Probabilistic Approach

This paper presents a data-driven approach for assessing the health of helicopter turbine engines, developed for the PHM North America 2024 Conference Data Challenge. The task involves both regression and classification to estimate the torque margin and classify engine health as either nominal or faulty. To quantify the reliability of predictions, probabilistic outputs are generated. We employ a two-stage model where the predicted torque margin serves as an input feature for health classification. For probabilistic torque margin estimation, we introduce an empirical error sampling method to generate torque margin samples, followed by a rule-based distribution selection scheme to evaluate the resulting distributions. For fault classification, logistic regression is used to provide confidence estimates, and we incorporate a score-optimized loss function during training to mitigate penalties for false negatives. Our approach demonstrates strong generalization to unseen assets, ranking 2nd in the competition with a score of 0.94, demonstrating its effectiveness in predicting health conditions and uncertainty for more informed helicopter engine management.

Advanced Disease Detection and Personalized Medicine: Integrated Data Approaches for Enhanced Parkinson's Disease and Breast Cancer Detection and Treatment in the USA

This present study focuses on integrating data-driven approaches into personalized medicine for better detection and treatment of Parkinson's disease and breast cancer through the US healthcare system. The deeper integration of genomics, clinical records, and patient self-reported data with machine learning algorithms will enhance early disease detection and optimization of treatment pathways. The results show that, more precisely, Random Forest and XGBoost machine learning models hold great promise for considerably improving diagnostic precision and predictive power. This realization opens a door for precision medicine-tailored health services according to the peculiarities of individual patients, which would improve treatment outcomes and encourage preventive healthcare. In addition, this approach aligns with the latest US efforts in precision medicine and contributes to evidence-based transformation in healthcare practice.

Advanced semantic segmentation of aircraft main components based on transfer learning and data-driven approach

Abstract The implementation of Smart Airport and Airport 4.0 visions relies on the integration of automation, artificial intelligence, data science, and aviation technology to enhance passenger experiences and operational efficiency. One essential factor in the integration is the semantic segmentation of the aircraft main components (AMC) perception, which is essential to maintenance, repair, and operations in aircraft and airport operations. However, AMC segmentation has challenges from low data availability, high-quality annotation scarcity, and categorical imbalance, which are common in practical applications, including aviation. This study proposes a novel AMC segmentation solution, employing a transfer learning framework based on a sophisticated DeepLabV3 architecture optimized with a custom-designed Focal Dice Loss function. The proposed solution remarkably suppresses the categorical imbalance challenge and increases the dataset variability with manually annotated images and dynamic augmentation strategies to train a robust AMC segmentation model. The model achieved a notable intersection over union of 84.002% and an accuracy of 91.466%, significantly advancing the AMC segmentation performance. These results demonstrate the effectiveness of the proposed AMC segmentation solution in aircraft and airport operation scenarios. This study provides a pioneering solution to the AMC semantic perception problem and contributes a valuable dataset to the community, which is fundamental to future research on aircraft and airport semantic perception. Graphical abstract