Mixed Data Research Articles

Joint Mixing Data Augmentation for Skeleton-based Action Recognition

Skeleton-based action recognition is beneficial for understanding human behavior in videos, and thus has received much attention in recent years as an important research area in action recognition. Current research focuses on designing more advanced algorithms to better extract spatio-temporal information from skeleton data. However, due to the small amount of data in the existing skeleton dataset and the lack of effective data augmentation methods, it is easy to lead to overfitting in model training. To address this challenge, we propose a mix-based data augmentation method, Joint Mixing Data Augmentation (JMDA), which can generally improve the effectiveness and robustness of various skeleton-based action recognition algorithms. In terms of spatial information, we introduce SpatialMix (SM), a method that projects the original 3D skeleton discrete information into a 2D space. Then, SM mixes the projected spatial information between two random samples during the training process to achieve the spatial-based mixing data augmentation. Concerning temporal information, we propose TemporalMix (TM). Leveraging the temporal continuity in skeleton data, we perform a temporal resize operation on the original skeleton data, and then merge two random samples during training to achieve the temporal-based mixed data augmentation. Additionally, we analyze the Feature Mismatch (FM) problem caused by introducing mix-based data augmentation into skeleton data. Then we propose a new data preprocessing method called Feature Alignment (FA) to effectively address this problem and improve model performance. Moreover, we propose a novel training pipeline, Joint Training Strategy (JTS), which combines multiple mix-based data augmentation methods for further improvement of model performance. Specifically, our proposed JMDA is plug-and-play and widely applicable to skeleton-based action recognition models. At the same time, the application of JMDA does not increase the model parameters and there is almost no additional training cost. We conduct extensive experiments on NTU RGB+D 60 and NTU RGB+D 120 datasets to demonstrate the effectiveness and robustness of the proposed JMDA on several mainstream skeleton-based action recognition algorithms.

Denoising diffusion probabilistic model enhanced tool condition monitoring method under imbalanced conditions

Abstract In recent years, tool condition monitoring (TCM) based on deep learning has been widely considered and achieved remarkable success. However, these methods typically require relatively large training samples to produce significant results, which are both imbalanced and rather troublesome to obtain in the practical application of TCM. To address this issue, a novel TCM method combined with multiscale permutation entropy (MPE), denoising diffusion probabilistic model (DDPM) and a residual network (ResNet) is proposed under conditions of sample imbalance. First, the one-dimensional sensing signal data is converted to a grayscale recurrence plot (RP) by minimizing the MPE of the signals in each channel. Second, combine and splice these grayscale RPs from different channels in each sample into color RPs. After that, the generated RP images using DDPM are added to the imbalanced dataset to augment the data to achieve a balanced state of the dataset. Finally, the balanced mixed data set of real and fake samples is input into a ResNet for recognition and monitoring tool conditions. TCM experiments are conducted to verify the performance of the proposed method with imbalanced dataset, and the results of experimental investigation demonstrate that the accuracy of the proposed method improved by 2%–18.8% compared to that of the other four sample augmentation methods using ResNet18 when the imbalance rate is 1:200.

Evaluating the therapeutic potential of different sources of mesenchymal stem cells in acute respiratory distress syndrome

BackgroundMesenchymal stem/stromal cells (MSCs) have attracted interest as a potential therapy given their anti-inflammatory and immunomodulatory properties. However, clinical trials using MSCs for acute respiratory distress syndrome (ARDS) have produced mixed and inconclusive data. In previous work, we performed a “head-to-head” comparison between different sources of MSCs and showed that each source had a unique genomic and proteomic “signature”.MethodThis study investigated which sources of MSC: bone marrow derived-MSCs (BM-MSCs), adipose tissue derived-MSCs (AD-MSCs) and umbilical cord derived-MSCs (UC-MSCs) would be the optimal candidate to be used as a therapy in an LPS-induced mouse model of ARDS. Immune cells assessment, tissue transcriptomics, animal survival, and endothelial-epithelial barrier assessment were used to evaluate their effects.ResultsWhen comparing the three most commonly used MSC sources, we found that UC-MSCs exhibited greater efficacy compared to other MSCs in improving animal survival, mitigating epithelial/endothelial damage, decreasing lung inflammation via reducing neutrophil infiltration, T cell proliferation, and M1 polarization. Bulk RNA sequencing of lung tissue also showed that UC-MSCs have the capability to downregulate extracellular trap formation, by the downregulation of key genes like Elane and Padi4. Notably, treatment with UC-MSCs demonstrated a significant reduction in Fc-γ R mediated phagocytosis, which has been associated with monocyte pyroptosis and intense inflammation in the context of COVID-19.ConclusionOur findings suggest that UC-MSCs are an optimal source of MSC to treat acute inflammatory conditions in the lungs, such as ARDS.

Attempt to Explore Ozone Mixing Ratio Data from Reanalyses for Trend Studies

In this paper, we use ozone mixing ratio data from the MERRA-2, ERA-5 and JRA-55 reanalyses from 500 hPa to 1 hPa in the period 1980–2020 with the aim of assessing their suitability for trend analysis. We found that these data are not suitable for trend studies due to huge differences in trend values and large differences in the variance of the ozone mixing ratio between reanalyses, and due to strong discrepancies between the ozone mixing ratio from reanalyses and that from the reliable ozonesonde at Hohenpeissenberg. These large differences can be caused by satellite replacement or by the assimilation of imperfect homogeneous data.

A recent shift in the Puccinia striiformis f. sp. tritici population in Serbia coincides with changes in yield losses of commercial winter wheat varieties

Wheat yellow (stripe) rust, caused by the fungus Puccinia striiformis f.sp. tritici (Pst), is a devastating disease of wheat worldwide. The success of Pst is largely due to the pathogens ability to rapidly overcome host resistance, generating new races that are easily dispersed between territories through wind-borne transmission of Pst urediniospores. Thus, first signs of entry of new Pst races into a region is usually captured by changes in disease severity. To examine any alterations of winter wheat variety response to Pst infection in Serbia, we analyzed yield and Pst disease severity in field trials conducted in 2014, 2021, and 2023. We specifically focused on analyzing Pst disease severity at growth stages related to yield. Associations between qualitative variables (variety, year) and quantitative variables (yield in untreated plots, yield loss, and disease index (DI) of Pst infection) were analyzed using Principal Component Analysis with mixed data. A General Linear Model was used to investigate the most influential factors on yield, yield loss, and Pst infection. The results indicated that yellow rust disease severity increased over the past decade, suggesting a potential recent change in the Pst population in Serbia. Comparative population genetic analysis of Pst samples from the 2023 wheat season and those collected in Serbia in 2014 confirmed a potential change in the Pst population. In addition, we found that yield losses across wheat varieties varied independently of Pst infection levels, indicating that wheat varieties differ in their ability to overcome damage caused by high levels of Pst infection. Given that the level of pathogen pressure triggering susceptibility reactions is cultivar-specific, our study highlights the need for a deeper focus on the mechanisms underlying these differences. Expanding our understanding of the interactions between pathogens, plant defense responses, and the ability of cultivars to mitigate yield losses will better equip us to predict and prevent potential yield losses in commercial wheat varieties due to yellow rust in the future.

Remnant cholesterol, clinical characteristics and mortality in 41.767 patients from LIPIDOGRAM 2004-2015 studies - the factor analysis for mixed data cluster analysis

Abstract Introduction Remnant cholesterol (remnant-C) contributes to residual cardiovascular risk and is produced by metabolism of triglyceride rich proteins. Aim To assess association between clinical characteristics of patients with elevated remnant-C in a large cohort of patients under the care of primary care physicians. Methods The LIPIDOGRAM studies were carried out in the primary care in Poland in 2004, 2006 and 2015. Patients (n=47,398) recruited in all 16 administrative regions in Poland and physicians were proportionally distributed to the number of inhabitants in each administrative region. Each patient was asked to fill the questionnaire on risk factors, chronic diseases, treatment and lifestyle. In the present analysis we included patients with body mass index (BMI )>18.5 kg/m2 aged 18-75 years. We set the cut-off point of fasting remnant cholesterol at >30 mg/dL (>0.77 mmol/L) that differentiated subjects at high risk of cardiovascular events. We carried out Factor Analysis for Mixed Data (FAMD) cluster analysis to discern groups of patients with similar clinical profiles. Results 41,767 patients, for which we had all relevant data, were included in the analysis. Follow-up rate was 97.8%. Three FAMD-derived clinical characteristics patterns accounted for 47.8% of the total variance and were retained for further analysis. The first pattern was associated with higher prevalence of metabolic syndrome (MetS), higher waist circumference, higher levels of non-HDL-C and remnant-C levels. The second pattern was distinguished by lipid parameters, particularly higher HDL-C levels, and was not significantly associated with comorbid conditions. The third pattern was linked to male sex, previous myocardial infarction, smoking, age, lower remnant-C levels, and reduced prevalence of obesity. All patterns were associated with 5-year mortality. The hazard ratio (HR) for mortality per 1 standard deviation (SD) increase in the first pattern score was 1.18 (95% CI: 1.16-1.22, p<0.001). Patients with clinical characteristics corresponding to second pattern had more favorable outcome HR (per 1SD score increase) – 0.75, 95%CI (0.73-0.78, p<0.001). Patients in third cluster, similarly to patients in cluster 1 had increased mortality HR (per 1SD score) – 1.10, (95%CI:1.06-1.15, p <0.001). Higher (>0.77 mmol/l / 30 mg/dl) remnant-C cholesterol significantly contributed to classification of patients into first and third clusters. It was positively associated with first pattern (r=0.64, p<0.001), but negatively with the third pattern (r=0.39, p<0.001). Conclusions Elevated Remnant-C was associated with clinical pattern typical for metabolic syndrome. Interestingly, lower remnant-C levels in patients with comorbidities and more advanced age may not automatically be indicative of a better prognosis – this requires further investigation.Figure 1.Figure 2.

The Impact of Seafarers' Welfare on the Operational Efficiency of Marine Vessels in Tanzania

This study explores the critical relationship between seafarers’ welfare and the operational efficiency of marine vessels in Tanzania. So the study aims to investigate the critical aspects of seafarers’ welfare and their impact on the operational efficiency of marine vessels in Tanzania. The objectives are, To identify the factors that affect the living and working conditions on the operational efficiency of a marine vessels inTanzania, To evaluate the working environment standard that influence the efficiency of a marine vessels, To examine the condition of their salary to afford the living standard. Tanzania ratified the MLC 2006, setting minimum international standards for seafarers' working and living conditions, significantly improving their welfare. Despite ratifying international conventions, Tanzanian seafarers face challenges on their welfare that the absence of a comprehensive maritime policy exacerbates these issues. This study employs a mixed methods research design, integrating both quantitative and qualitative data collection and analysis to provide a comprehensive understanding of the impact of seafarers’ welfare on the operational efficiency of marine vessels. A sample size of 80 participants from various marine vessels and ship managers were selected regression and thematic analysis was been used in data analysis and concurrent triangulation method was used in data intergration. The findings of this study underscore the importance of prioritizing seafarers’ welfare to ensure the sustainable and efficient operation of marine vessels in Tanzania. By addressing the identified welfare issues. The research recommend for policy changes and the development of a more supportive and productive maritime sector as well as establishment of welfare centers like Stella Maris and mision to seafarers, to support seafarers and fishing communities.

Imputation methods for mixed datasets in bioarchaeology

Missing data is a prevalent problem in bioarchaeological research and imputation could provide a promising solution. This work simulated missingness on a control dataset (481 samples × 41 variables) in order to explore imputation methods for mixed data (qualitative and quantitative data). The tested methods included Random Forest (RF), PCA/MCA, factorial analysis for mixed data (FAMD), hotdeck, predictive mean matching (PMM), random samples from observed values (RSOV), and a multi-method (MM) approach for the three missingness mechanisms (MCAR, MAR, and MNAR) at levels of 5%, 10%, 20%, 30%, and 40% missingness. This study also compared single imputation with an adapted multiple imputation method derived from the R package “mice”. The results showed that the adapted multiple imputation technique always outperformed single imputation for the same method. The best performing methods were most often RF and MM, and other commonly successful methods were PCA/MCA and PMM multiple imputation. Across all criteria, the amount of missingness was the most important parameter for imputation accuracy. While this study found that some imputation methods performed better than others for the control dataset, each imputation method has advantages and disadvantages. Imputation remains a promising solution for datasets containing missingness; however when making a decision it is essential to consider dataset structure and research goals.

Human-AI Collaboration: A Student-Centered Perspective of Generative AI Use in Higher Education

The increasing adoption of generative AI (Gen AI) has made it even more important to investigate how education and learning is transformed. The paper investigates the evolving relationship between humans and AI in performing learning and knowledge-intensive tasks. Based on mixed data collected from business school students, the article explores the evolving relationships between human and AI in knowledge collaboration. The article sets out to address how Gen AI usage affects students’ behavior, their academic work and their attitudes towards AI. It investigates how students use Gen AI in an academic context to support students’ work processes. Our analysis draws on collaborative learning theories and revisits the automation-augmentation paradox in the context of management education. The aim is to address the following research questions: How does collaboration with AI affect student academic work? And how do students in higher education use Gen AI for academic work? We present insights into students’ perception of benefits and drawbacks of using Gen AI, their attitude towards Gen AI, knowledge tasks in which they collaborate or delegate to Gen AI and we discuss the potential risks associated with the use of Gen AI. Students are collaborating with AI using ChatGPT for expanding knowledge on academic themes, summarizing concepts, theories, and generating ideas for research topics and methods. Time saving, enhanced productivity and user-friendliness of the tool were identified as the main benefits associated with Gen AI, whereas the risk of plagiarism, its inaccuracy in responses and the need for prior knowledge were identified as the main drawbacks. While our findings underscore the potential of Gen AI to significantly enhance student learning experiences, they also underscore the importance of exercising caution and awareness of associated risks to automate learning. This study seeks to enrich our comprehension of AI's transformative role in higher education, with a specific focus on the student-centered perspective.

Artificial intelligence can be trained to predict c-KIT-11 mutational status of canine mast cell tumors from hematoxylin and eosin-stained histological slides.

Numerous prognostic factors are currently assessed histologically and immunohistochemically in canine mast cell tumors (MCTs) to evaluate clinical behavior. In addition, polymerase chain reaction (PCR) is often performed to detect internal tandem duplication (ITD) mutations in exon 11 of the c-KIT gene (c-KIT-11-ITD) to predict the therapeutic response to tyrosine kinase inhibitors. This project aimed at training deep learning models (DLMs) to identify MCTs with c-KIT-11-ITD solely based on morphology. Hematoxylin and eosin (HE) stained slides of 368 cutaneous, subcutaneous, and mucocutaneous MCTs (195 with ITD and 173 without) were stained consecutively in 2 different laboratories and scanned with 3 different slide scanners. This resulted in 6 data sets (stain-scanner variations representing diagnostic institutions) of whole-slide images. DLMs were trained with single and mixed data sets and their performances were assessed under stain-scanner variations (domain shifts). The DLM correctly classified HE slides according to their c-KIT-11-ITD status in up to 87% of cases with a 0.90 sensitivity and a 0.83 specificity. A relevant performance drop could be observed when the stain-scanner combination of training and test data set differed. Multi-institutional data sets improved the average accuracy but did not reach the maximum accuracy of algorithms trained and tested on the same stain-scanner variant (ie, intra-institutional). In summary, DLM-based morphological examination can predict c-KIT-11-ITD with high accuracy in canine MCTs in HE slides. However, staining protocol and scanner type influence accuracy. Larger data sets of scans from different laboratories and scanners may lead to more robust DLMs to identify c-KIT mutations in HE slides.

The feasibility and acceptability of engaging older adults living with multiple long-term conditions, frailty, and a recent deterioration in health in research: Findings from the Lifestyle in Later Life – Older People’s Medicine (LiLL-OPM) study

BackgroundOlder adults living with multiple long-term conditions (MLTC, also known as multimorbidity) and frailty are more likely to experience a deterioration in their health requiring specialist referral or hospital admission than individuals without these syndromes. However, this group of older people are underserved by research meaning that there is a limited evidence base for their care. This study therefore aimed (1) to determine if it is feasible to recruit and collect quantitative data to describe the health and lifestyle of older adults living with MLTC, frailty and a recent deterioration in health and (2) to assess if taking part in research is acceptable to this group of older adults.MethodsParticipants were approached and recruited for this study via an Older People’s Medicine Day Unit in Newcastle upon Tyne, UK. The study took a mixed methods approach, involving quantitative and qualitative data collection. To determine the feasibility of carrying out research in this group, we quantified recruitment rate and collected data on the health and lifestyle, including diet and physical activity, of the participants. Qualitative semi-structured interviews were undertaken to assess acceptability. Two separate interviews were carried out focusing on involving older adults in research and the participants’ experiences of taking part in the research. Interviews were analysed using thematic analysis.ResultsFifty patients were approached to participate in the study with twenty-nine (58%) successfully recruited. It was feasible to collect information to describe the health and lifestyle of these older adults who demonstrated very low levels of physical activity. Participants reported that taking part in the research was acceptable to them with interview analysis generating three themes (1) developing a meaningful partnership, (2) enabling factors to participation: research at home with flexible delivery and (3) social and psychological benefits of research participation.ConclusionsIt is feasible and acceptable to recruit and carry out research with this underserved group of older adults. Participants found taking part in this research to be acceptable and reported overall positive experiences of their involvement in the study and indicated that they would be willing to contribute to further research in the future.

Socio-Economic Status of Individuals in Tehran University of Medical Sciences Employees` Cohort Study Using PCA, MCA and FAMD Methods

Introduction: Determining socio-economic status (SES) can greatly help decision makers in the field of social health. Because SES can play an important role in accessing medical services or welfare amenities. We aimed to determine the SES using principal component analysis (PCA), multiple correspondence analysis (MCA), and factor analysis of mixed data (FAMD) methods. Methods: In this cross-sectional study (2023), 4448 employees aged 19 to 75 years were included to the study from Tehran University of Medical Sciences employees` cohort (TEC). Demographic variables and socioeconomic factors were considered. Considering the weaknesses of PCA and MCA methods, we calculated the SES score using PCA, MCA and FAMD methods, and the percentile of people was determined. These weaknesses include normality assumption and considering only linear relationship for PCA, inability to interpret the relationships between variables and considering each level of classification variables as a new variable for MCA. Results: We studied 4448 people (39.3% men) with mean age of 42.3 and a standard deviation of 8.7.The correlation between the percentiles obtained through PCA, MCA and FAMD methods was very high, and the highest correlation was related to the percentiles obtained through PCA and FAMD methods with a value of 0.994. The intraclass correlation coefficient value was 0.996. Also, this value was 0.996 and 0.994 in the random samples of 250 and 100 individuals from the original data, respectively. Conclusion: All of the three methods worked similarly on determining the SES and calculating the percentile of people. PCA and FAMD methods had better agreement than others. Therefore, in studies that have both quantitative and qualitative variables, the choice of analysis method depends on the opinion of the researcher.

Accurate and 30-plus days reliable cuffless blood pressure measurements with 9-minutes personal photoplethysmograph data and mixed deduction learning

Cuffless blood pressure (BP) measurements have long been anticipated, and the PPG (Photoplethysmography)-only method is the most promising one since already embedded in many wearable devices. To further meet the clinical accuracy requirements, PPG-only BP predictions with personalized modeling for overcoming personal deviations have been widely studied, but all required tens to hundreds of minutes of personal PPG measurements for training. Moreover, their accurate test periods without calibration practice were not reported. In this work, we collected records of PPG data from our recruited subjects in real-life scenarios instead of relying on the openly available MIMIC dataset obtained from intensive care unit (ICU) patients. Since our objective is commercial application and a substantial reduction in training data, we tailored our model training to closely mimic real-world usage. To achieve this, we developed a training approach that only requires 9-minutes of personal PPG signal recordings and mixed with other PPG data from our recruited 364 subjects. The modeling is conducted with two-channel paired inputs to the convolutional neural network (CNN)-based model, which we called Mixed Deduction Learning (MDL). The test results of 88 samples from 15 subjects, under testing period up to 30-plus days without extra calibration, revealed that MDL meets most of the standards of AAMI, BHS, and IEEE 1708–2014 (for static test only) for BP measurement devices, which indicates MDL’s long-term stability and consistency. Furthermore, we found that the model with two-channel inputs presents a trend of improving performance as the pool of mixed training data increased, while the conventional one-channel input revealed degraded performance. The outperformance of MDL is attributed to many significant features remained in the first CNN layer even when mixing personal 9-minutes data with the other 364 subjects. Consequently, PPG-only with MDL introduces a new avenue for overcoming challenges in training due to personal physiological variations. Given our consideration of real-life usage, this technology can be seamlessly translated to commercial applications.

Adaptive Online Continual Learning for In-Situ Quality Prediction in Manufacturing Processes

Abstract Manufacturing processes undergo continuous changes in order to meet various requirements, such as process/product changes and variations in tool/workpiece conditions, leading to mixed, heterogenous, or anomalous data. As a result, a quality prediction model trained from previous data may not perform well when new tasks emerge. In order to achieve in-time and accurate product quality prediction, it is crucial to develop a predictive method that adapt to variations in the manufacturing system, capable of learning from new tasks without forgetting previous ones and detecting unknown tasks. This study proposes a deep learning method integrated with continual learning for in-situ quality prediction that is capable of learning from new tasks without forgetting previous ones. To demonstrate this idea, deep CNNs are designed to analyze in-process sensor data, which consist of shared layers to capture the common underlying features across all tasks, and task-specific layers that capture specific characteristics of each individual task. In order to identify the task to which incoming product belongs, a task prediction approach based on task relevancy using filter subspace distance is proposed. When new data come in, the model first identifies the task, followed by predicting the quality of the current product. The proposed method is demonstrated on two case studies, including quality prediction of the workpiece using acoustic emissions during the laser-induced plasma micro-machining process, and quality prediction of the product through thermal images during the hot stamping process.

Exploring Multisource High‐Dimensional Mixed‐Frequency Risks in the Stock Market: A Group Penalized Reverse Unrestricted Mixed Data Sampling Approach

Exploring Multisource High‐Dimensional Mixed‐Frequency Risks in the Stock Market: A Group Penalized Reverse Unrestricted Mixed Data Sampling Approach

Comparative Impact of Virtual and Face-To-Face Education on the Academic Performance and Perception of Accounting Students at a Peruvian University during the Pandemic

The present study compares the impact of virtual and face-to-face education modalities on the academic performance and perception of accounting students at the Universidad Nacional Toribio Rodríguez de Mendoza de Amazonas during the COVID-19 pandemic. Through a mixed methodological approach, quantitative and qualitative data from 98 students were analyzed, evaluating their academic performance, interaction, feedback and satisfaction in both modalities. The results show that students in the virtual modality obtained higher academic performance, attributed to the flexibility of online education. However, interaction and feedback were perceived as more effective in the face-to-face modality. Students valued the flexibility of virtual education, but emphasized the importance of personal interaction and the traditional classroom environment to maintain concentration and academic engagement. This study concludes that both modalities have strengths and weaknesses, and suggests that a hybrid combination could offer a more complete educational experience. Improvements in feedback and assessment in virtual environments are recommended, as well as encouraging active participation through interactive technologies.

Intuitive-K-prototypes: A mixed data clustering algorithm with intuitionistic distribution centroid

Data sets are usually mixed with numerical and categorical attributes in the real world. Data mining of mixed data makes a lot of sense. This paper proposes an Intuitive-K-prototypes clustering algorithm with improved prototype representation and attribute weights. The proposed algorithm defines intuitionistic distribution centroid for categorical attributes. In our approach, a heuristic search for initial prototypes is performed. Then, we combine the mean of numerical attributes and intuitionistic distribution centroid to represent the cluster prototype. In addition, intra-cluster complexity and inter-cluster similarity are used to adjust attribute weights, with higher priority given to those with lower complexity and similarity. The membership and non-membership distance are calculated using the intuitionistic distribution centroid. These distances are then combined parametrically to obtain the composite distance. The algorithm is judged for its clustering effectiveness on the real UCI data set, and the results show that the proposed algorithm outperforms the traditional clustering algorithm in most cases.

A wrapper feature selection approach using Markov blankets

In feature selection, Markov Blanket (MB) based approaches have attracted considerable attention with most MB discovery algorithms being categorized as filter based techniques. Typically, the Conditional Independence (CI) test employed by such methods is different for different data types. In this article, we propose a novel Markov Blanket based wrapper feature selection method. The proposed approach employs Predictive Permutation Independence (PPI), a novel Conditional Independence (CI) test that allows it to work out-of-the-box for both classification and regression tasks on mixed data. PPI can work with any supervised algorithm to estimate the association of a feature with the target variable while also providing a measure of feature importance. The proposed approach also includes an optional MB aggregation step that can be used to find the optimal MB under non-faithful conditions. Our method11Implementation and experimental results are available at https://github.com/AnonymousMLSubmissions/PatternRecognitionSubmission. outperforms other MB discovery methods, in terms of F1-score, by 7% on average, over 3 large-scale BN datasets. It also outperforms state-of-the-art feature selection techniques on 13 real-world datasets.

Mixed fuzzy C-means clustering

Clustering analysis becomes challenging when the dataset has mixed data types comprising categorical (nominal or ordinal scale) and numerical (interval scale) features. Mainstream distance metrics cannot handle the information in categorical data about the similarity between the observations and cluster centers, leading to performance loss. Various methods are introduced in the literature to handle the mixed data types in clustering. However, each method has disadvantages in capturing categorical information about similarity, adjusting the contribution of categorical information to clustering, and computational or implementation inefficiency. This study proposes a mixed fuzzy C-means clustering method for mixed data types. Two new distance metrics are developed to handle binary and multi-class nominal features. The scaled entropy of each data type is used to adjust the weight of each data type in the overall similarity metric, providing a lower bias since no user-specified weight is required. A comparative numerical study is conducted with twenty real datasets and seven benchmark methods using five cluster validation statistics. The mixed fuzzy C-means clustering performs better than the benchmark methods and is computationally efficient in practice. Since all the computer codes for implementing mixed fuzzy C-means clustering are given, the proposed method is readily applicable to practical problems.

Integrating Mixup and ArcFace for enhanced anomalous sound detection

In machine condition monitoring, accurate detection of anomalous sounds is essential for timely maintenance interventions. Traditional deep neural network models employ classification-based self-supervised learning to discriminate normal and abnormal operational sounds using classification confidence. However, these models often suffer from overfitting or excessive confidence, limiting their effectiveness. To overcome this, we introduce a hybrid technique that combines the strengths of Mixup and ArcFace techniques. Mixup enhances model generalization by training on mixed sound data, while ArcFace promotes greater inter-class distances with an angular margin loss. However, there is a challenge in combining these methods because the angular margin for ArcFace is defined for a single target class, and hence, can be hardly defined for the mixed data utilized in Mixup. The proposed technique employs the class-dependent angular margins based on the mixing ratio of data. SoftMax losses for individual classes are constructed using the class-dependent margins and then integrated into the Mixup loss. When applied to the DCASE2020 Task2 dataset, the proposed method significantly outperforms the state-of-the-art NoisyArcMix model in anomaly detection. This approach demonstrates the potential of leveraging class-dependent angular margins in mixed data for improved performance in machine condition monitoring.