Categorization Method Research Articles

How to best assess shedder status: a comparison of popular shedder tests.

"Shedder status" describes the inherent variation between individuals to leave touch DNA on a surface through direct contact. Depending on the amount and quality of DNA or cellular deposition, individuals are typically deemed high, intermediate, or low shedders. Although many shedder tests have been described, variability in study design and categorisation criteria has limited the ability of researchers to accurately compare results, as well as accrue the necessary population data. As activity level reporting becomes more common, the need for reliable and standardised testing increases. To assess reproducibility, this study compared shedder status data generated by six participants using three different shedder tests, as modified from the literature. This involved DNA quantification and profiling of a handprint made on a glass plate, DNA quantification and profiling of a grip mark made on a plastic conical tube, and cell scoring of a Diamond™ Dye-stained fingermark. All participants washed and dried their hands fifteen minutes before each deposit. To assess the impact of behaviour on shedder designation, participants either refrained from activity or went about their daily tasks during this wait. The shedder status of participants changed between tests, as DNA-based testing often generated lower shedder statuses than cell scores. Further, when different categorisation methods were applied to a single test, intra-person variability increased as the number of shedder designations increased from two (low/high) to five (low/low-intermediate/intermediate/intermediate-high/high). Moving forward, the utilisation of a single shedder test and standardised categorisation criteria is needed to employ shedder testing in forensic casework.

Accelerometer-Based Pavement Classification for Vehicle Dynamics Analysis Using Neural Networks

This research examines the influence of various pavement types on vehicle dynamics, specifically concentrating on vertical acceleration and its implications for unsprung mass, including the wheels and suspension system. The objective of this project was to categorize pavement types with accelerometer data, enabling a deeper comprehension of the impact of road surface conditions on vehicle stability, comfort, and mechanical stress. Two categorization methods were utilized: a neural network and a multinomial logistic regression model. Accelerometer data were gathered while a car navigated diverse terrain types, such as grates, potholes, and cobblestones. The neural network model exhibited exceptional performance, with 100% accuracy in categorizing all surface types, while the multinomial logistic regression model reached 97.14% accuracy. The neural network demonstrated exceptional efficacy in differentiating intricate surface types such as potholes and grates, surpassing the logistic regression model which had difficulties with these surfaces. These results underscore the neural network’s effectiveness in the real-time categorization of road surfaces, enhancing the comprehension of vehicle dynamics influenced by pavement conditions. Future studies must tackle the difficulty of identifying analogous surfaces by enhancing methodologies or integrating more data attributes for greater precision.

An Enhanced Affective Computing Technology for Fostering an Emotionally Healthy Workplace

The mental health of workers is of paramount importance in today’s fast-paced and demanding workplaces. This research establishes a strong connection between affective decision-making and mental health, presenting a novel method to improve affective computing technologies for developing emotionally healthy workplaces. We use Multinomial Naive Bayes Integrated Gated Recurrent Units (MNB-GRU) for classification and prediction to reach our goals. This dataset includes many measures of mental health, working climate, and individual variables. Data Exploration is performed to learn more about the properties of the dataset, and Preprocessing Grouping is used to get the data ready for analysis. Relationships between emotional decisionmaking and mental health markers are shown using data visualization approaches to give intuitive insights. To evaluate the reliability of the connection, a correlation analysis is used in Model Assessment. Understanding how people are feeling emotionally at work can be gained via this assessment, which examines the degree of association between affective decision-making and mental health. As a result of using categorization methods to divide the workforce into several categories based on their emotional well-being, we can better assist businesses in meeting the varying demands of their staff members. This method guarantees that efforts to improve mental health are focused and productive. By creating a solid connection between effective decisionmaking and mental health, businesses can take preventative measures to aid their workers' emotional well-being and create a more upbeat and productive workplace.

Deciphering the Symptom Spectrum: A Comprehensive Analysis of Migraine Patterns and Types

Complex neurological diseases like migraine migraine affect a large section of the global population, causing health, social, and economic issues. Migraine causes intense, painful headaches that are usually one-sided and pulsing. Auras, nausea, vomiting, and excessive light and sound sensitivity may precede these episodes. Migraine affect millions worldwide and can be intermittent or persistent, impairing function. Diet and stress may induce it, but the cause is unknown. Prevention and symptom treatment drugs and lifestyle changes are used. Debilitating migraines are hard to diagnose due to their varied presentation and subjective symptom reporting. Traditional migraine diagnosis, based on clinical evaluation, typically fails to classify migraine types, requiring more objective and rigorous instruments. This study proposes a machine learning-based migraine categorization method to address this issue. The dataset includes different patient demographics and clinical variables; thus, we use complex algorithms like Random for Forest, XGBoost, and Extra Trees. These algorithms are great for deciphering migraine patterns because they excel at evaluating complex datasets. The research seeks to close this gap to improve migraine classification accuracy, objectivity, and reliability, enabling tailored migraine management and treatment. This neurology study could im- prove migraine diagnosis and treatment with more effective and personalized plans.

Leveraging Explainable Artificial Intelligence (XAI) for Expert Interpretability in Predicting Rapid Kidney Enlargement Risks in Autosomal Dominant Polycystic Kidney Disease (ADPKD)

Autosomal dominant polycystic kidney disease (ADPKD) is the predominant hereditary factor leading to end-stage renal disease (ESRD) worldwide, affecting individuals across all races with a prevalence of 1 in 400 to 1 in 1000. The disease presents significant challenges in management, particularly with limited options for slowing cyst progression, as well as the use of tolvaptan being restricted to high-risk patients due to potential liver injury. However, determining high-risk status typically requires magnetic resonance imaging (MRI) to calculate total kidney volume (TKV), a time-consuming process demanding specialized expertise. Motivated by these challenges, this study proposes alternative methods for high-risk categorization that do not rely on TKV data. Utilizing historical patient data, we aim to predict rapid kidney enlargement in ADPKD patients to support clinical decision-making. We applied seven machine learning algorithms—Random Forest, Logistic Regression, Support Vector Machine (SVM), Light Gradient Boosting Machine (LightGBM), Gradient Boosting Tree, XGBoost, and Deep Neural Network (DNN)—to data from the Polycystic Kidney Disease Outcomes Consortium (PKDOC) database. The XGBoost model, combined with the Synthetic Minority Oversampling Technique (SMOTE), yielded the best performance. We also leveraged explainable artificial intelligence (XAI) techniques, specifically Local Interpretable Model-Agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP), to visualize and clarify the model’s predictions. Furthermore, we generated text summaries to enhance interpretability. To evaluate the effectiveness of our approach, we proposed new metrics to assess explainability and conducted a survey with 27 doctors to compare models with and without XAI techniques. The results indicated that incorporating XAI and textual summaries significantly improved expert explainability and increased confidence in the model’s ability to support treatment decisions for ADPKD patients.

Collagen Type VII (COL7A1) as a Longevity Mediator in Caenorhabditis elegans: Anti-Aging Effects on Healthspan Extension and Skin Collagen Synthesis

Longevity genes and senescence-related signaling proteins are crucial targets in aging research, which aims to enhance the healthy period and quality of life. Identifying these target proteins remains challenging because of the need for precise categorization and validation methods. Our multifaceted approach combined bioinformatics with transcriptomic data to identify collagen as a key element associated with the lifespan of the model organism, Caenorhabditis elegans. By analyzing transcriptomic data from long-lived mutants that involved mechanisms such as antioxidation, dietary restriction, and genetic background, we identified collagen as a common longevity-associated gene. We validated these findings by confirming that collagen peptides positively affect lifespan, thereby strengthening the validity of the target. Further verification through healthspan factors in C. elegans and functional assays in skin fibroblasts provided additional evidence of the role of collagen in organismal aging. Specifically, our study revealed that collagen type VII is a significant target protein for mitigating age-related decline. By validating these findings across different aging models and cell-based studies, we present compelling evidence for the anti-aging effects of collagen type VII, highlighting its potential as a target for promoting healthy aging. This study proposes that collagen not only serves as an indicative marker of organismal longevity across various senescence-related signaling pathways, but also offers a mechanistic understanding of skin degeneration. Consequently, collagen is an effective target for interventions aimed at mitigating skin aging. This study underscores the potential of collagen type VII (bonding collagen T7) as a therapeutic target for enhancing skin health and overall longevity.

Clinical features of vascular cognitive impairment

Vascular cognitive impairment (VCI) can occur in cerebrovascular disease of varying degree of severity, which requires further investigation.Objective: to identify specific features of attention deficit and executive function (EF) impairment in VCI and to determine the relationship between cognitive, neurological and radiological signs of the disease.Material and methods. The study involved 80 patients (51 with VCI – the main group and 29 healthy volunteers – the control group). All participants underwent clinical and neuropsychological examination and MRI of the brain. VCI was diagnosed according to 2014 VASCOG criteria. The neuropsychological examination included an assessment of the general cognitive function (MoCA test, MMSE); general EF (FAB, EXIT-25); episodic memory (12-word memory test and Benton Visual Retention Test); semantic memory (categorical association method); initiation and productivity of cognitive activity (literal associations); attention [TMT-A, Symbol-Digit Modalities Test (SDMT)]; cognitive flexibility (TMT-B); cognitive control (Stroop test).Results. As a result of the study, three subgroups of patients with VCI were identified: 1) predominantly attention deficit (AD); 2) AD and significant decrease in cognitive flexibility (CFD); 3) AD in combination with a significant cognitive control deficiency (CCD). The TMT-A and SDMT tests were significantly worse in all subgroups of VCI compared to the control group, with the worst SDMT score in subgroup 2. Subgroup 2 also had the longest time to complete the TMT-B test. Subgroup 3 was characterized by the worst Stroop test results, and uncontrolled arterial hypertension was more common in this group. Gait disturbances were significantly more frequent in subgroup 2, and increased reflexes of oral automatism and grasp reflex were observed in subgroup 3. White matter hyperintensity (WMH) was more pronounced in subgroup 2, and lacunar infarcts (LI) were more frequent in subgroup 3. Prevalence of LI and severity of WMH were lowest in subgroup 1.Conclusion. The main features of VCI are attention deficit and EF impairment, characteristics of EF disorders depend on the severity and type of vascular lesions of the brain. Progression of WMH is associated with more pronounced CFD and an increase in the number of LI is associated with CCD. There is a correlation between CFD and gait disturbances and between CCD and lack of primitive reflexes inhibition.

The Use of Digital Technologies in the Administration of Justice: Positive and Negative Effects

The subject of the study is social relations in the field of the use of digital technologies in the administration of justice. The purpose of the study is to identify and describe the positive and negative effects of the use of information technology in the administration of justice in their interrelation and interdependence. The hypothesis of the study assumes the existence of a set of interrelated and interdependent effects of the use of digital technologies that affect the administration of justice in different ways (positively and negatively). The categorical and symbolic method “hexagram” is applied. The most significant positive effects of the use of digital technologies in the administration of justice are: realization of the human rights; access to justice; optimization of the work of the court. The most significant negative effects of the use of digital technologies in the administration of justice are: digital inequality, risks of information security violation, the risk of losing the individualization of court decisions. The obtained result can be used in the selection of priority areas for the implementation of digital technologies in the administration of justice, as well as in the planning of this activity, prevention of risks of violation of human rights or principles of justice in connection with the use of digital technologies.

Machine Learning based Suggestion Method for Land Suitability Assessment and Production Sustainability

The global population is projected to increase by an additional two billion by 2050, as per the assessment conducted by Food and Agriculture Management. However, the arable land is anticipated to expand by just 5%. Consequently, intelligent and effective agricultural practices are essential to enhancing farming production. Evaluating rural Land Suitability (LS) is a crucial instrument for agricultural growth. Numerous novel methods and concepts are being adopted in agriculture as alternatives for gathering and processing farm data. The swift advancement of wireless Sensor Networks (WSN) has prompted the creation of economical and compact sensor gadgets, with the Internet of Things (IoT) serving as a viable instrument for automation and decision-making in farmers. To evaluate agricultural LS, this study offers an expert system integrating networked sensors with Machine Learning (ML) technologies, including neural networks. The suggested approach would assist farmers in evaluating agricultural land for cultivating across four decision categories: very appropriate, suitable, somewhat suitable, and inappropriate. This evaluation is based on the data gathered from various sensor devices for system training. The findings achieved with the MLP with four concealed layers demonstrate efficacy for the multiclass categorization method compared to other current models. This trained system will assess future evaluations and categorize the land post-cultivation.

RERconverge Expansion: Using Relative Evolutionary Rates to Study Complex Categorical Trait Evolution.

Comparative genomics approaches seek to associate molecular evolution with the evolution of phenotypes across a phylogeny. Many of these methods lack the ability to analyze non-ordinal categorical traits with more than two categories. To address this limitation, we introduce an expansion to RERconverge that associates shifts in evolutionary rates with the convergent evolution of categorical traits. The categorical RERconverge expansion includes methods for performing categorical ancestral state reconstruction, statistical tests for associating relative evolutionary rates with categorical variables, and a new method for performing phylogeny-aware permutations, "permulations", on categorical traits. We demonstrate our new method on a three-category diet phenotype and we compare its performance to binary RERconverge analyses and two existing methods for comparative genomic analyses of categorical traits: phylogenetic simulations and a phylogenetic signal based method. We present an analysis of how the categorical permulations scale with the number of species and the number of categories included in the analysis. Our results show that our new categorical method outperforms phylogenetic simulations at identifying genes and enriched pathways significantly associated with the diet phenotypes and that the categorical ancestral state reconstruction drives an improvement in our ability to capture diet-related enriched pathways compared to binary RERconverge when implemented without user input on phenotype evolution. The categorical expansion to RERconverge will provide a strong foundation for applying the comparative method to categorical traits on larger data sets with more species and more complex trait evolution than have previously been analyzed.

A Comparative Analysis Machine Learning Techniques for LULC Classification Using Landsat-8 Satellite Imagery

This study compares various categorization methods to assign land use and land cover (LULC) classes. Using Geographic Information Systems (GIS) and Remote Sensing (RS) to leverage the dynamic and complex area of LULC, this study examines the potential of different machine learning classification methods. Precise differentiation and classification of various land cover categories, such as green vegetation, urban areas, water bodies, dark green vegetation, and bare terrain, are made possible by the great spatial and spectral resolution of Landsat imagery. For efficient land management and planning, the integration of Landsat data with GIS and RS approaches provides insightful information about the distribution and temporal changes in LULC. This study uses four classifiers to explore the principles of supervised machine learning techniques and identify their drawbacks and advantages. Testing results show that the Support Vector Machine with four kernel linear-99.17%, radial basis (RBF)-99.11%, Sigmoid-99.11% and Polynomial-99.11% is a reliable option for LULC classification, outperforming than other classifiers in terms of accuracy, including the Minimum Distance Classifier (MD-93.47%), Maximum Likelihood Classifier (MLC- 98.98%), and Mahalanobis Distance Classifier (MH-97.83%). Among the tested classifiers, SVM with four kernels notably shows the highest accuracy. With their essential insights for well-informed decision-making towards sustainable development and resource utilization, our findings add to a thorough understanding of LULC dynamics. For accurate mapping and long-term monitoring of deviations in land cover (LC), the study emphasizes the value of using front-line classification systems in remote sensing applications.

Evaluating Students' Academic Progress in the Role of Fuzzy Logic

This paper explores the role of fuzzy logic in the assessment of student performance in local high school, Badr, Libya, positing that it offers a more flexible and accurate alternative to traditional categorization methods of academic achievement. Conventional assessment tools, such as standardized test scores and GPAs, often yield a distorted and overly simplistic view of student performance. Utilizing a mixed-methods research design, this study compares the performance, attendance, and self-reported satisfaction of students assessed through fuzzy logic with those evaluated by traditional methods. The results demonstrate that students assessed using fuzzy logic achieved higher average scores and expressed greater satisfaction with the assessment process. Additionally, educators employing fuzzy logic provided more detailed and constructive feedback, enriching the learning experience. Thus, this research indicates that fuzzy logic can effectively mitigate the shortcomings of traditional assessment methods, enhancing the recognition of diverse student abilities. The findings advocate for the integration of fuzzy logic in educational assessment practices to promote inclusivity and improved learning outcomes.

A Relative Projection-Based Multiattribute Group Decision-Making Model With Noncooperative Behavior Management and Application to NEV Supplier Selection.

This article reports a novel consensus model where a group of internal and external experts evaluate alternatives under multiple attributes and provide mutual evaluations. First, different from previous studies, the cognitive and interest conflicts of internal and external experts are considered simultaneously. But interest conflict is emphasized for internal experts, and cognitive conflict is mainly considered for external experts. Second, we explore the categorization and management methods of noncooperative behaviors (NCBs) of experts. The relative projection-based indexes are proposed for the first time to measure the degrees of cognitive and interest conflicts by using multiattribute preference matrices (MAPMs) and the weight vectors of attributes. A group of experts are divided into three categories and the corresponding management strategies are developed. Third, we investigate the consensus mechanism among experts with cognitive and interest conflicts. For reaching an acceptable consensus level, an adjustment process is proposed to revise some local entries in MAPMs and mutual evaluation matrix (MEM). A penalty mechanism is further established to dynamically update the weights of experts. An algorithm is designed to capture the consensus reaching process in multiattribute group decision making, where internal and external experts are distinguished by proposing a parameter. Finally, the high-performance battery supplier selection of new energy vehicle is studied to illustrate the proposed model. The results reveal that the efficiency of reaching consensus can be enhanced by using the developed model with effective management of NCBs.

디자인 씽킹을 활용한 주민참여 지역재생 사업 사례 조사 - 부여군 사례를 중심으로

This study aims to revitalize local commercial districts amidst the challenges of the COVID era by applying the design thinking methodology to develop a regeneration strategy for “Wonjo Food Alley” in Buyeo County. The research was conducted based on workshops involving merchants, local residents, and experts, utilizing the KJ method for idea categorization and analysis to derive a strategy for community regeneration using the cultural assets of the area. In this process, the use of design thinking played a crucial role in deeply understanding the unique requirements of the region and facilitating community participation. The results demonstrate that design thinking is an effective approach for revitalizing local economies, proving its utility and effectiveness. This underscores the importance of a participatory approach through design thinking in the development of strategies for commercial district revitalization, and it is hoped that this research will provide practical guidelines for addressing similar problems in the future.

Who is a minor? Age assessments of refugees in Germany and the classificatory multiplicity of the state

The categories that states use to classify and govern migrants often diverge from how migrants themselves perceive and experience their conditions, biographies, and identities. Building on this insight, I argue in this article that state categories not only clash with migrants’ varied notions of categories: state categorization itself often relies on multiple definitions of categories and methods of categorization. Age assessments of young refugees in Germany demonstrate this multiplicity. These assessments are carried out by forensic medical examiners and social workers, respectively, who work with very different understandings of what constitutes age. While forensic scientists view age as countable, particular, and objective, social workers think of it as lived, assembled, and constructed. What I term the classificatory multiplicity of the state complements current scholarly endeavors aimed at critically examining state categories, particularly in the context of migration governance.

Quality of birth care and risk factors of length of stay after birth: A machine learning approach.

Length of stay (LOS) is an outcome measure and is assumed to be related to quality. The objective of this study is to examine the quality of birth care and risk factors associated with LOS after birth. A nationwide population-based Turkish Demographic and Health Survey (TDHS) was used for the year 2018. A total of 1849 women ages 15-49 were included. Explanatory factor analysis and machine learning predictors such as Random Forest, Support Vector Machine, Neural Network, k-Nearest Neighbor, and Naïve Bayes were used to identify the quality of birth care and risk factors associated with LOS after birth. As a result of the explanatory factor analysis, factor structures of quality of birth care, antenatal check-ups and supplements, and risk factors associated with birth were obtained using the Categorical Component Analysis method. The type of delivery, place of delivery, age, and type of place, which are under the quality of birth care, and risk factors associated with birth factors were found to be the variables that had the highest impact on LOS estimation. Random forest (Accuracy = 0.5789), support vector machine (radial) (Accuracy = 0.5766), and neural network (Accuracy = 0.5750) models outperformed, respectively. Type of delivery which is an indicator of quality of birth care is a strong predictor of LOS after birth according to the Random Forest model. We demonstrated that machine learning techniques offer precise LOS prediction after birth. Further studies assessing the effect of quality of birth care on predicting LOS at birth would be beneficial.

Circuit categorization approach of office building energy consumption based on data features for energy-saving diagnosis

Energy-saving in office buildings is crucial. Research on data-driven diagnostics of building energy consumption is pivotal. The emergence and development of itemized electricity platforms have expanded the scope of diagnosis from total building energy consumption to individual electrical equipment. However current studies rarely address the characteristics of sub-circuits. This paper highlights the importance of recognizing the features of electrical equipment sub-circuits based on historical energy consumption data by examining various aspects of energy use in office buildings, including electrical consumption disaggregation, equipment operational strategies, sub-item consumption proportion, and sub-circuit composition. A circuit categorization approach in office buildings by features analysis of historical data is innovatively proposed. Operational conditions, time series stationarity, and correlations with factors affecting energy consumption are considered in historical data feature analysis. The method categorizes the electric circuits into 5 categories with typical features based on the ADF-KPSS co-test and Spearman correlation analysis. The rationality and validity of the circuit categorization method were verified with actual building data in case studies. On this basis, an interpretable energy consumption anomaly diagnosis method based on the categorization results is proposed. The circuit categorization approach explores how to further investigate the typical features of sub-circuits, and provides new directions for subsequent building operation and maintenance(O&M) research, including energy consumption diagnostics. Additionally, it offers decision support for building managers for O&M evaluation from an interpretable perspective.

Combining machine learning (ML) and participatory rural appraisal (PRA) for disaster risk preparedness (DRP): Evidence from the poorest region of Luzon, Philippines

In the field of social science, disaster risk preparedness (DRP) is considered immeasurable due to its multidimensional nature, making it infamously difficult to quantify. The current measurements are costly, labor-intensive, and time-consuming. Consequently, policymakers struggle to target policies effectively when implementing disaster risk reduction management initiatives. By combining Participatory Rural Appraisal (PRA) and Machine Learning (ML) to train and test community-based system datasets, this work proposes novel approaches to DRP in the poorest region of Luzon, Philippines. We utilized sophisticated econometrics models along with ML categorization methods. Through the analysis of 34 locales and 4 sectors within a disaggregation system over 429 ensemble runs using cross-validation techniques, we then combined the results. The Support Vector Machine (SVM) classifier achieved the highest accuracy of 91.55 % randomly and 94.53 % within the pipeline, surpassing all other models. It also confirms the current relationship between DRP and multidimensional attributes (a total of 21 factors) in terms of correlation and causation. Our work showcases the potential of ML for disaster risk prediction, potentially reducing costs, saving labor, and optimizing time, especially in the most impoverished areas of the Philippines. Ultimately, through extensive PRA, the outcomes have provided different localities with tools for targeting policies in disaster risk management.

The efficacy of Diamond™ nucleic acid dye-stained cell counting techniques for forensic application

Touch DNA is one of the most common types of biological material collected during criminal investigations. Diamond™ Nucleic Acid Dye (DD) has been shown to aid in touch sample visualisation and target sampling. It has also been used as a method of shedder categorisation that is cheaper and quicker than DNA methods. However, the DD method routinely involves manual cell counting, which can result in intra and inter-person variability similar to other manual techniques used in forensic science, for example, fingerprint identification. Additionally, DD based shedder categorisation involves counting cells in a portion of the touch deposit to extrapolate an individual’s shedder status, and the sampling effect of such estimations is currently unknown.The present study tested different data analysis aspects of the DD method, including counting variability within and between people, shedder classification differences based on different counting methods (entire thumbprint, sub-section of a print with most cells, sub-section of a print deemed most representative of the entire thumbprint, and random sections), the use of ImageJ software to semi-automate counting and the use and extension of the DD method for investigating DNA Transfer, Persistence, Prevalence and Recovery (DNA-TPPR).The results of this study show that there are meaningful differences observed during counting processes both between and within people. These differences tended to increase as the factor of time, or the duration of counting, rather than the complexity of cell deposits being assessed. Investment in cell counting software that eliminates personal factors, such as boredom fatigue, can remedy most of these issues, however, will require optimisation, such as fibre recognition. Shedder testing was shown to be affected by the choice of sampling and categorisation methods, and suggested that using an entire finger or larger section size can provide increased precision. Finally, inverted worn gloves stained with DD may provide an acceptable alternative for hands in DNA-TPPR investigations, providing an interesting alternative for future research.

Assessment of Accelerated Aging Effect of Bio-Oil Fractions Utilizing Ultrahigh-Resolution Mass Spectrometry and k-Means Clustering of van Krevelen Compositional Space.

Bio-oils contain a substantial number of highly oxygenated hydrocarbons, which often exhibit low thermal stability during storage, handling, and refining. The primary objectives of this study are to characterize the hydroxyl group in bio-oil fractions and to investigate the relationship between the type of hydroxyl group and accelerated aging behavior. A bio-oil was fractionated into five solubility-based fractions, classified in two main groups: water-soluble and water-insoluble fractions. These fractions were then subjected to chemoselective reactions to tag molecules containing hydroxyl groups and analyzed by negative-ion electrospray ionization 21 T Fourier transform ion cyclotron resonance mass spectrometry (FT-ICR MS). The fractions were also subjected to accelerated aging experiments and characterized by FT-ICR MS and bulk viscosity measurements. Extracting insightful information from ultrahigh-resolution data to aid in predicting upgrading methodologies and instability behaviors of bio-oils is challenging due to the complexity of the data. To address this, an unsupervised learning technique, k-means clustering analysis, was used to semiquantify molecular compositions with a close Euclidean distance within the (O/C, H/C) chemical space. The combination of k-means analysis with findings from chemoselective reactions allowed the distinctive hydroxyl functionalities across the samples to be inferred. Our results indicate that the hexane-soluble fraction contained numerous molecules containing primary and secondary alcohols, while the water-soluble fraction displayed diverse groups of oxygenated compounds, clustered near to carbohydrate-like and pyrolytic humin-like materials. Despite its high oxygen content, the water-soluble fraction showed minimal changes in viscosity during aging. In contrast, a significant increase in viscosity was observed in the water-insoluble materials, specifically, the low- and high-molecular-weight lignin fractions (LMWL and HMWL, respectively). Among these two fractions, the HMWL exhibited the highest increase in viscosity after only 4 h of accelerated aging. Our results indicate that this aging behavior is attributed to an increased number of molecular compositions containing phenolic groups. Thus, the chemical compositions within the HMWL are the major contributors to the viscosity changes in the bio-oil under accelerated aging conditions. This highlights the crucial role of oxygen functionality in bio-oil aging, suggesting that a high oxygen content alone does not necessarily correlate with an increase of viscosity. Unlike other bio-oil categorization methods based on constrained molecule locations within the van Krevelen compositional space, k-means clustering can identify patterns within ultrahigh-resolution data inherent to the unique chemical fingerprint of each sample.