Hierarchical Class Research Articles

Information-extreme machine learning of an ophthalmic diagnostic system with a hierarchical class structure

The paper considers the method of hierarchical information-extreme machine learning for the system of ophthalmic diagnosis of eye pathology. Since the proposed method is developed within the framework of a functional approach to modeling the cognitive processes of natural intelligence, it, unlike neuro-like structures, acquires the properties of flexibility when retraining a diagnostic system and requires an order of magnitude fewer image samples. In addition, the decision rules based on the results of machine learning within the geometric approach in the form of a binary hierarchical structure of recognition classes ensure their practical invariance to the multidimensionality of both the space of diagnostic features and the alphabet of recognition classes. The modified Kullback-Leibler information measure, which is considered as a function of the accuracy of classification solutions, is chosen as a criterion for optimizing the parameters of the machine learning system for diagnosing eye pathologies. A hierarchical information-extreme machine learning algorithm for an ophthalmic diagnostic system for six eye pathologies was developed and programmatically implemented. Based on the results of functional diagnostics, it has been experimentally proved that the constructed decision rules are error-free according to the training matrices of recognition classes of each level of the constructed binary hierarchical structure.

Exploring the influence of attention for whole-image mammogram classification

Attention is an important component of modern Convolutional Neural Networks (CNNs) that has been shown to improve baseline model performance for a wide variety of tasks. Attention has shown specific promise in the classification and segmentation of mammograms, but we have a limited understanding of why attention improves performance in these domains. In this paper, we present a robust comparison of different combinations of baseline models and attention methods at two resolutions for whole mammogram classification of masses and calcifications. We find that attention generally helps to improve baseline model performance. However, the extent of improvement is governed by a combination of model architecture and the statistical characteristics of the data. Specifically, we show that high amounts of pooling and model complexity may result in decreased performance for data with high variability. To better understand the effect of attention on mammogram classification, we used LayerCAM, a hierarchical Class Activation Map (CAM) approach, to visualize where the network pays attention in the input image. This research provides statistical evidence that attention can improve the correlation between model performance and LayerCAM activation in the region of interest (ROI). However, these correlations are weak and variable, indicating that improvements in model performance due to attention are not necessarily caused by increased model activation near the ROI. Overall, our work provides novel insights to help guide future efforts in incorporating attention-based mechanisms for mammogram classification.

Understanding Reasons for Vaccination Hesitancy and Implementing Effective Countermeasures: An Online Survey of Individuals Unvaccinated against COVID-19.

This online survey of unvaccinated people living in Japan aimed to identify the reasons for declining vaccination and to develop effective countermeasures. We conducted a hierarchical class analysis to classify participants, examine factors influencing their classification, and provide the information they needed about coronavirus disease 2019 (COVID-19) and trusted sources of COVID-19 information for each group. A total of 262 participants were classified into three groups: Group 1 with no specific reason (28 participants, 10.69%); Group 2 with clear concerns about trust in the vaccine (85 participants, 32.44%), and Group 3 with attitudinal barriers, such as distrust of the vaccine and complacency towards COVID-19, and structural barriers, such as vaccination appointments (149 participants, 56.87%). For each group, females tended to be classified in Group 2 more than Group 1 (Odds ratio (OR) [95% confidential intervals (95%CI)] = 1.64 (0.63 to 2.66), p = 0.001) and in Group 3 more than Group 1 (OR [95%CI] = 1.16 (0.19 to 2.12), p = 0.019). The information that the participants wanted to know about COVID-19 was different among each group (Safety: p < 0.001, Efficacy: p < 0.001, Genetic effects: p < 0.001). Those who did not receive the COVID-19 vaccine also had lower influenza vaccination coverage (8.02%). Additionally, 38 participants (14.50%) were subject to social disadvantages because they had not received the COVID-19 vaccine. Countermeasures should be carefully tailored according to the target population, reasons for hesitancy, and specific context. The findings of this study may help develop individualized countermeasures to address vaccine hesitancy.

BNVGLENET: Hypercomplex Bangla handwriting character recognition with hierarchical class expansion using Convolutional Neural Networks

Object recognition technology has made significant strides where recognizing handwritten Bangla characters including symbols, compounds form, etc. remains a challenging problem due to the prevalence of cursive writing and many ambiguous characters. The complexity and variability of the Bangla script, and individual’s unique handwriting styles make it difficult to achieve satisfactory performance for practical applications, and the best existing recognizers are far less effective than those developed for English alpha-numeric characters. In comparison to other major languages, there are limited options for recognizing handwritten Bangla characters. This study has the potential to improve the accuracy and effectiveness of handwriting recognition systems for the Bengali language, which is spoken by over 200 million people worldwide. This paper aims to investigate the application of Convolutional Neural Networks (CNNs) for recognizing Bangla handwritten characters, with a particular focus on enlarging the recognized character classes. To achieve this, a novel challenging dataset for handwriting recognition is introduced, which is collected from numerous students’ handwriting from two institutions. A novel convolutional neural network-based approach called BNVGLENET is proposed in this paper to recognize Bangla handwritten characters by modifying the LeNet-5 and combining it with the VGG architecture, which has the advantage of significantly identifying the characters from Bengali handwriting. This study systematically evaluated the performance of models not only on custom novel dataset but also on the publicly available Bangla handwritten character dataset called the Grapheme dataset. This research achieved a state-of-the-art recognition accuracy of 98.2% on our custom testing vowel-consonant class and 97.5% on the custom individual class. The improvements achieved in this study bridge a notable disparity between the practical needs and the actual performance of Bangla handwritten character recognition systems.

Interpretable multi-domain meta-transfer learning for few-shot fault diagnosis of rolling bearing under variable working conditions

To address the challenges of accurately diagnosing few-shot fault samples obtained from rolling bearings under variable operating conditions, as well as the issues of black box nature and delayed feedback to guide fault handling in intelligent diagnostic models, this paper proposes an interpretable multi-domain meta-transfer learning method. Firstly, vibration monitoring data of rolling bearings under different operating conditions are collected, and time–frequency domain features are extracted to construct multi-channel one-dimensional temporal samples as inputs. A multi-domain meta-transfer learning framework based on deep convolutional neural networks is then built to perform few-shot learning with multiple tasks under different operating conditions. The output results are reverse-reconstructed through a fusion hierarchical class activation mapping, and the feature maps are assigned different weights to obtain saliency maps corresponding to the inputs, thus improving the interpretability of the output results. Finally, the dataset of bearing vibration data under time-varying rotational speed conditions is used to validate the effectiveness of the proposed method. Experimental results show that the proposed method can achieve accurate fault diagnosis results under variable operating conditions with few-shot samples, and the diagnosis results can be fed back to the input for decision-making, enhancing the interpretability of the model. Compared with other models, it also demonstrates better robustness and accuracy.

Boosting crop classification by hierarchically fusing satellite, rotational, and contextual data

Accurate early-season crop type classification is crucial for the crop production estimation and monitoring of agricultural parcels. However, the complexity of the plant growth patterns and their spatio-temporal variability present significant challenges. While current deep learning-based methods show promise in crop type classification from single- and multi-modal time series, most existing methods rely on a single modality, such as satellite optical remote sensing data or crop rotation patterns. We propose a novel approach to fuse multimodal information into a model for improved accuracy and robustness across multiple crop seasons and countries. The approach relies on three modalities used: remote sensing time series from Sentinel-2 and Landsat 8 observations, parcel crop rotation and local crop distribution. To evaluate our approach, we release a new annotated dataset of 7.4 million agricultural parcels in France (FR) and the Netherlands (NL). We associate each parcel with time-series of surface reflectance (Red and NIR) and biophysical variables (LAI, FAPAR). Additionally, we propose a new approach to automatically aggregate crop types into a hierarchical class structure for meaningful model evaluation and a novel data-augmentation technique for early-season classification. Performance of the multimodal approach was assessed at different aggregation levels in the semantic domain, yielding to various ranges of the number of classes spanning from 151 to 8 crop types or groups. It resulted in accuracy ranging from 91% to 95% for the NL dataset and from 85% to 89% for the FR dataset. Pre-training on a dataset improves transferability between countries, allowing for cross- domain and label prediction, and robustness of the performances in a few-shot setting from FR to NL, i.e., when the domain changes as per with significantly new labels. Our proposed approach outperforms comparable methods by enabling deep learning methods to use the often overlooked spatio-temporal context of parcels, resulting in increased precision and generalization capacity.

Activity phenomena in the baryonic universe as a result of interaction between baryonic objects and the carrier of dark energy

Non-stable phenomena are under consideration for rather long time. Mostly two issues seem to be very essential in connection with non-stability, namely, their role in the evolution of cosmic objects and the sources providing energy for non-stability manifestations. We argue here that the energy exchange with the dark energy carrier may be the main mechanism of energy providing. Backed by the contemporary ideas about the dark energy, we accept that this new-found energy homogeneously fills all space at all scales and interacts with cosmic objects belonging to all hierarchical levels. We argue that baryons mass is changeable quantity depending on the environmental physical conditions and therefore it is a subject of evolution as the objects of other hierarchical classes. Moreover, we argue in favour of new paradigm to be used for the further development of cosmology and cosmogony, which is not based on the a priory hypothesis by Kant and Laplace.

КАТЕГОРИЯ НАБЛЮДАТЕЛЯ В РОМАНЕ: ОПЫТ МЕТОДОЛОГИЧЕСКОЙ РЕФЛЕКСИИ

The article outlines the contour of understanding the methodological principles underlying the strategy of researcher’s scientific activity in line with the methodology of functional pragmatism. It analyzes the structure of scientific concepts and clarifies the scope and content of the terms “concept” and “category” as well as the hierarchical class structure (intensional) of the categorical concept “observer”, which allows to distinguish it from related concepts, such as: visual in literature, artistic whole, direct and reverse perspective, point of view, subject of vision, speech, evaluation, and others. In addition, detailed comments are given for each of the hierarchical classes of the categorical structure of the concept “observer”, which are successively generalized into a single definition

Evolution of Human Races and the advent of Aryans in South Asia

This study investigates the Aryan race's origins, migration into India, and their transformative impact on South Asian civilization. Utilizing a qualitative approach, the research synthesizes secondary sources, including historical texts, scholarly articles, and archaeological findings, to provide a detailed account of Aryan influence. The findings reveal that Aryans migrated into India between 500 and 2500 BC, interacting with various indigenous groups like the Austric and Darwin races. Their arrival marked a significant shift from simpler, egalitarian societies based on agriculture and animal husbandry to a more complex social structure with hierarchical class systems and centralized governance. The Aryans introduced new administrative practices and social norms, laying the groundwork for a more structured society. Furthermore, the Aryans played a crucial role in the establishment and promotion of Hinduism, incorporating ethical principles from religious texts into their governance and cultural practices. This integration fostered the development of enduring Hindu festivals, rituals, and traditions that continue to be celebrated today. The research underscores the Aryans' pivotal role in shaping India's cultural and societal landscape. It provides valuable insights for historians and scholars of South Asian studies, enhancing the understanding of early human migration, cultural integration, and the formation of ancient civilizations in the Indian subcontinent.

CSTS: Exploring Class-Specific and Task-Shared Embedding Representation for Few-Shot Learning.

Few-shot learning (FSL) is a challenging yet promising technique that aims to discriminate objects based on a few labeled examples. Learning a high-quality feature representation is key with few-shot data, and many existing models attempt to extract general information from the sample or task levels. However, the common sample-level means of feature representation limits the models generalizability to different tasks, while task-level representation may lose class characteristics due to excessive information aggregation. In this article, we synchronize the class-specific and task-shared information from the class and task levels to obtain a better representation. Structure-based contrastive learning is introduced to obtain class-specific representations by increasing the interclass distance. A hierarchical class structure is constructed by clustering semantically similar classes using the idea of granular computing. When guided by a class structure, it is more difficult to distinguish samples in different classes that have similar characteristics than those with large interclass differences. To this end, structure-guided contrastive learning is introduced to study class-specific information. A hierarchical graph neural network is established to transfer task-shared information from coarse to fine. It hierarchically infers the target sample based on all samples in the task and yields a more general representation for FSL classification. Experiments on four benchmark datasets demonstrate the advantages of our model over several state-of-the-art models.

Sentic-Emotion Classifier on eWallet Reviews

Emotion classification using hybrid framework using lexicon and machine learning algorithms have been proven to be more accurate. This research analyses emotions from reviews of a popular eWallet mobile application in Malaysia. The proposed Sentic-Emotion Classifier is evaluated on its performance as it analyses the code-switched reviews crawled that contain formal and informal or out-of-vocab words. The code-switched reviews are mainly made up of words and expressions in English and Malay language models. This research designs, implements, and investigates several novel techniques that have been shown to have reliable and consistent predictive outcomes, and these outcomes are validated with manually annotated reviews so that the proposed classifier can be evaluated objectively. The novel contributions of the Sentic-Emotion Classifier consist of 2-tier sentiment classification, extended emolex framework, and multi-layer discrete emotion hierarchical classes which is hypothesized to be able to yield better accuracy for emotion and intensity prediction for the proposed framework.

Intra-operative applications of augmented reality in glioma surgery: a systematic review.

Augmented reality (AR) is increasingly being explored in neurosurgical practice. By visualizing patient-specific, three-dimensional (3D) models in real time, surgeons can improve their spatial understanding of complex anatomy and pathology, thereby optimizing intra-operative navigation, localization, and resection. Here, we aimed to capture applications of AR in glioma surgery, their current status and future potential. A systematic review of the literature was conducted. This adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline. PubMed, Embase, and Scopus electronic databases were queried from inception to October 10, 2022. Leveraging the Population, Intervention, Comparison, Outcomes, and Study design (PICOS) framework, study eligibility was evaluated in the qualitative synthesis. Data regarding AR workflow, surgical application, and associated outcomes were then extracted. The quality of evidence was additionally examined, using hierarchical classes of evidence in neurosurgery. The search returned 77 articles. Forty were subject to title and abstract screening, while 25 proceeded to full text screening. Of these, 22 articles met eligibility criteria and were included in the final review. During abstraction, studies were classified as "development" or "intervention" based on primary aims. Overall, AR was qualitatively advantageous, due to enhanced visualization of gliomas and critical structures, frequently aiding in maximal safe resection. Non-rigid applications were also useful in disclosing and compensating for intra-operative brain shift. Irrespective, there was high variance in registration methods and measurements, which considerably impacted projection accuracy. Most studies were of low-level evidence, yielding heterogeneous results. AR has increasing potential for glioma surgery, with capacity to positively influence the onco-functional balance. However, technical and design limitations are readily apparent. The field must consider the importance of consistency and replicability, as well as the level of evidence, to effectively converge on standard approaches that maximize patient benefit.

Experimental Observations of the Topology of Convolutional Neural Network Activations

Topological data analysis (TDA) is a branch of computational mathematics, bridging algebraic topology and data science, that provides compact, noise-robust representations of complex structures. Deep neural networks (DNNs) learn millions of parameters associated with a series of transformations defined by the model architecture resulting in high-dimensional, difficult to interpret internal representations of input data. As DNNs become more ubiquitous across multiple sectors of our society, there is increasing recognition that mathematical methods are needed to aid analysts, researchers, and practitioners in understanding and interpreting how these models' internal representations relate to the final classification. In this paper we apply cutting edge techniques from TDA with the goal of gaining insight towards interpretability of convolutional neural networks used for image classification. We use two common TDA approaches to explore several methods for modeling hidden layer activations as high-dimensional point clouds, and provide experimental evidence that these point clouds capture valuable structural information about the model's process. First, we demonstrate that a distance metric based on persistent homology can be used to quantify meaningful differences between layers and discuss these distances in the broader context of existing representational similarity metrics for neural network interpretability. Second, we show that a mapper graph can provide semantic insight as to how these models organize hierarchical class knowledge at each layer. These observations demonstrate that TDA is a useful tool to help deep learning practitioners unlock the hidden structures of their models.

Hierarchical few-shot learning with feature fusion driven by data and knowledge

Few-shot learning (FSL) aims to use only a few samples to learn a model and utilizes that model to identify unseen classes. Recent, metric-based feature fusion methods mainly focus on the fusion of inter-layer features and show superior performance in solving FSL problems. However, due to the data scarcity in FSL, existing methods still face severe challenges in obtaining high-quality sample features for the improvement of classification performance. In this paper, we propose a hierarchical metric FSL model with comprehensive feature fusion driven by data and knowledge (HFFDK), which is based on intra-layer channel-feature and hierarchical class structure perspectives. First, we utilize the network hierarchy to construct an intra-layer channel feature fusion, which transfers the intra-layer fused features of the higher layer to the lower layer. The model can extract high-quality sample features in a data-driven manner. Moreover, we focus on different levels of granularity to obtain various levels of information, while hierarchical class structures can provide both coarse- and fine-grained information in a knowledge-driven manner. Then, we utilize the coarse-grained information to assist fine-grained recognition. Finally, we optimize hierarchical FSL with coarse- and fine-grained relational constraints and similarity measures among samples. Experiments on four benchmark datasets show that HFFDK achieves state-of-the-art performance.

Multiclass classification of Fermi-LAT sources with hierarchical class definition

ABSTRACT In this paper, we develop multiclass classification of Fermi-large area telescope (LAT) gamma-ray sources using machine learning with hierarchical determination of classes. One of the main challenges in the multiclass classification of the Fermi-LAT sources is that the size of some of the classes is relatively small, for example with less than 10 associated sources belonging to a class. In this paper, we propose a hierarchical structure for the determination of the classes. This enables us to have control over the size of classes and to compare the performance of the classification for different numbers of classes. In particular, the class probabilities in the two-class case can be computed either directly by the two-class classification or by summing probabilities of children classes in multiclass classification. We find that the classifications with few large classes have comparable performance with classifications with many smaller classes. Thus, on one hand, the few-class classification can be recovered by summing probabilities of classification with more classes while, on the other hand, the classification with many classes gives a more detailed information about the physical nature of the sources. As a result of this work, we construct three probabilistic catalogues, which are available online. This work opens up a possibility to perform population studies of sources including unassociated sources and to narrow down searches for possible counterparts of unassociated sources, such as active galactic nuclei, pulsars, or millisecond pulsars.

D. H. Lawrence’ın Lady Chatterley’in Sevgilisi Romanında Üst ve Alt Kültür İkilemi

Lawrence, in Lady Chatterley’s Lover (1928), reflects the post-war modern British society which condemns its individuals into a psychological trauma of ongoing conflicts between different classes and cultures. Advocating the exigency of a classless society, Lawrence shatters the hegemonic ideology of the highbrow culture which disdains working class culture as vulgar, coarse and commonplace. Reacting against the cultural displacement of working classes who are considered as masses, Lawrence indulges in revealing the authenticity and earnestness of the working class culture. Lawrence’s depiction of an uncorrupt ethical standing of the working class culture is the negation of the concept of culture, associated with high class manners and ways of life. The prevailing concern of this study, therefore, will be to analyze Lawrence’s Lady Chatterley’s Lover from the perspective of cultural studies to unravel the clash of high and low cultures in relation to the hierarchical class structure of the post-war England.

Building hierarchical class structures for extreme multi-class learning

Building hierarchical class structures for extreme multi-class learning

Learning EKG Diagnostic Models with Hierarchical Class Label Dependencies.

Electrocardiogram (EKG/ECG) is a key diagnostic tool to assess patient's cardiac condition and is widely used in clinical applications such as patient monitoring, surgery support, and heart medicine research. With recent advances in machine learning (ML) technology there has been a growing interest in the development of models supporting automatic EKG interpretation and diagnosis based on past EKG data. The problem can be modeled as multi-label classification (MLC), where the objective is to learn a function that maps each EKG reading to a vector of diagnostic class labels reflecting the underlying patient condition at different levels of abstraction. In this paper, we propose and investigate an ML model that considers class-label dependency embedded in the hierarchical organization of EKG diagnoses to improve the EKG classification performance. Our model first transforms the EKG signals into a low-dimensional vector, and after that uses the vector to predict different class labels with the help of the conditional tree structured Bayesian network (CTBN) that is able to capture hierarchical dependencies among class variables. We evaluate our model on the publicly available PTB-XL dataset. Our experiments demonstrate that modeling of hierarchical dependencies among class variables improves the diagnostic model performance under multiple classification performance metrics as compared to classification models that predict each class label independently.

Salmonella spp. in Aquaculture: An Exploratory Analysis (Integrative Review) of Microbiological Diagnoses between 2000 and 2020.

The present study aimed to characterize, through descriptive statistics, data from scientific articles selected in a systematic integrative review that performed a microbiological diagnosis of Salmonella spp. in aquaculture. Data were obtained from research articles published in the BVS, Scielo, Science Direct, Scopus and Web of Science databases. The selected studies were published between 2000 and 2020 on samples of aquaculture animal production (fish, shrimp, bivalve mollusks, and other crustaceans) and environmental samples of aquaculture activity (farming water, soil, and sediments). After applying the exclusion criteria, 80 articles were selected. Data such as country of origin, categories of fish investigated, methods of microbiological diagnosis of Salmonella spp., sample units analyzed and most reported serovars were mined. A textual analysis of the word cloud and by similarity and descending hierarchical classification with the application of Reinert's algorithm was performed using R® and Iramuteq® software. The results showed that a higher percentage of the selected articles came from Asian countries (38.75%). Fish was the most sampled category, and the units of analysis of the culture water, muscle and intestine were more positive. The culture isolation method is the most widespread, supported by more accurate techniques such as PCR. The most prevalent Salmonella serovars reported were S. Typhimurium, S. Weltevreden and S. Newport. The textual analysis showed a strong association of the terms "Salmonella", "fish" and "water", and the highest hierarchical class grouped 25.4% of the associated text segments, such as "aquaculture", "food" and "public health". The information produced characterizes the occurrence of Salmonella spp. in the aquaculture sector, providing an overview of recent years. Future research focusing on strategies for the control and prevention of Salmonella spp. in fish production are necessary and should be encouraged.

The structure of self-concept and mental well-being in university students in China: A HiCLAS analysis

• The relationship between self-concept and mental well-being has been well-researched in the literature. However, the use of variables that consider the contextual background of the samples causes the findings incomparable. • Using simple-structural self-concept variables derived from the Hierarchical Classes Analyses (HICLAS), we found that a more differentiated self is associated with more depressive symptoms, lower resilience, and lower self-esteem. Self-esteem plays a role as a mediator between the relationship of self-concept with depressive symptoms and resilience. • Our observations inform future interventions that target self-concept for youth mental health.