Hierarchical Classification Research Articles

Six classes named entity recognition for mapping location of Indonesia natural disasters from twitter data

PurposeThe purpose of this study is to provide the location of natural disasters that are poured into maps by extracting Twitter data. The Twitter text is extracted by using named entity recognition (NER) with six classes hierarchy location in Indonesia. Moreover, the tweet then is classified into eight classes of natural disasters using the support vector machine (SVM). Overall, the system is able to classify tweet and mapping the position of the content tweet.Design/methodology/approachThis research builds a model to map the geolocation of tweet data using NER. This research uses six classes of NER which is based on region Indonesia. This data is then classified into eight classes of natural disasters using the SVM.FindingsExperiment results demonstrate that the proposed NER with six special classes based on the regional level in Indonesia is able to map the location of the disaster based on data Twitter. The results also show good performance in geocoding such as match rate, match score and match type. Moreover, with SVM, this study can also classify tweet into eight classes of types of natural disasters specifically for the Indonesian region, which originate from the tweets collected.Research limitations/implicationsThis study implements in Indonesia region.Originality/value(a)NER with six classes is used to create a location classification model with StanfordNER and ArcGIS tools. The use of six location classes is based on the Indonesia regional which has the large area. Hence, it has many levels in its regional location, such as province, district/city, sub-district, village, road and place names. (b) SVM is used to classify natural disasters. Classification of types of natural disasters is divided into eight: floods, earthquakes, landslides, tsunamis, hurricanes, forest fires, droughts and volcanic eruptions.

“When I’m a Human Being”: race, bodies, and power in Disney’s adaptations of “The Frog King or Iron Henry”

In the summer of 2020, Walt Disney Theme Parks announced plans to reimagine the popular attraction “Splash Mountain” in the theme of the 2009 feature-length animated film The Princess and the Frog. As only the third Disney feature-length animated film to showcase Black characters prominently, The Princess and the Frog seems perfectly poised to replace the Song of the South—critiqued for its romanticization of race relations after the American Civil War—in the Disney parks. In this adaptation of “Der Froschkönig oder der eiserne Heinrich” (“The Frog King or Iron Henry”), Walt Disney Animation Studios introduces its first Black princess. While a hardworking, self-starter princess of color is a step towards representation, the film’s portrayal of non-white characters reinforces a hierarchy of race and class which privileges the wealthy white characters while relegating characters of color to the bottom rungs of human society and into non-human bodies. This essay explores the intersection of gender, race, and class in Disney’s adaptations of the classic fairy tale “Der Froschkönig” from the Grimms’ 1812 variation and the plans to reimagine it for Disney theme parks. In conversation with adaptation studies, narrative theory, and feminist theories, this contribution interrogates the underlying messages of prejudice and -isms in The Princess and the Frog that reinforce twentieth- and twenty-first-century prejudices against people of color, women, and the working class, rendering the proposed update of the “Splash Mountain” attraction into a modernization of the problems already recognized in the current ride.

Assessment of surface water quality in a highly anthropised semi-arid catchment (Case of Wadi El Hatab Basin, Tunisia, North Africa)

Anthropogenic contamination of surface waters is a critical issue in urbanised catchments, especially in Mediterranean environments with scarce water resources and intermittent rivers. A combined hydro-geochemical and multivariate statistical methodology has been applied to evaluate the surface water quality of Wadi El Hatab catchment (Central Tunisia). More than half of the 28 samples collected in March 2021 exceeded WHO standards for sodium, chloride and sulphate contents. Most polluted sites are directly influenced by domestic and industrial effluents that contaminate the river for kilometers. Principal Component Analysis and Hierarchical Ascendant Classification distinguished four contamination sources: water-rock interaction, domestic wastewater, untreated industrial effluents mainly from paper mills and irrigation return. A Water Quality Index model indicated that 75% of the surface water was unsuitable for drinking purposes; while according to chemical indices such as saturation index, sodium absorption coefficient (SAR), magnesium ratio (MR) and percent sodium (%Na), 25 to 49% of samples were unsuitable for irrigation. Considering that water of Wadi El Hatab is used for vegetable irrigation, serious public health issues may arise. The investigation demonstrated that a single representative sampling campaign could be a useful way to identify contamination sources when interpreted with both statistical and geochemical tools.

RNAVirHost: a machine learning-based method for predicting hosts of RNA viruses through viral genomes.

The high-throughput sequencing technologies have revolutionized the identification of novel RNA viruses. Given that viruses are infectious agents, identifying hosts of these new viruses carries significant implications for public health and provides valuable insights into the dynamics of the microbiome. However, determining the hosts of these newly discovered viruses is not always straightforward, especially in the case of viruses detected in environmental samples. Even for host-associated samples, it is not always correct to assign the sample origin as the host of the identified viruses. The process of assigning hosts to RNA viruses remains challenging due to their high mutation rates and vast diversity. In this study, we introduce RNAVirHost, a machine learning-based tool that predicts the hosts of RNA viruses solely based on viral genomes. RNAVirHost is a hierarchical classification framework that predicts hosts at different taxonomic levels. We demonstrate the superior accuracy of RNAVirHost in predicting hosts of RNA viruses through comprehensive comparisons with various state-of-the-art techniques. When applying to viruses from novel genera, RNAVirHost achieved the highest accuracy of 84.3%, outperforming the alignment-based strategy by 12.1%. The application of machine learning models has proven beneficial in predicting hosts of RNA viruses. By integrating genomic traits and sequence homologies, RNAVirHost provides a cost-effective and efficient strategy for host prediction. We believe that RNAVirHost can greatly assist in RNA virus analyses and contribute to pandemic surveillance.

О КЛАССИФИКАЦИИ МЕЛИОРАТИВНЫХ НАСОСНЫХ СТАНЦИЙ

Purpose: analysis and construction of hierarchical and facet classifications of reclamation pumping stations (RPS). During the research, a search, analysis, comparison and systematization of RPS types were carried out in accordance with regulatory and technical documentation, as well as the developments of various Russian scientists. When developing and constructing RPS classifications, the following methods were used: facet and hierarchical. Discussion. As a result of the study and analysis of various sources of information, it was found that the existing RPS classifications in most cases are presented according to one classification criterion. In this regard, a classification has been carried out at several levels. First, the main classification criteria such as the purpose and main characteristics, including flow and pressure, were determined, according to which the hierarchical classification was built. Subsequently, lower order classification features were considered: importance classes, degree of durability, design, type of building, structure configuration, location relative to the irrigation source, location in the system, degree of use, type of water intake structure and source, devices used and control principle. Based on the indicated classification criteria, hierarchical and facet classifications of the RPS were formed. Conclusions. As a result of research, a hierarchical classification of the RPS by main types with a hierarchy of vague subordination up to the seventh level with minor groups of classification characteristics has been developed. The use of the facet method made it possible to combine all types and kinds of RPS into a more detailed classification with the introduction of some independent classification features of lower levels, specifying certain types of RPS within the hierarchical classification levels. The developed hierarchical and facet classifications of RPS make it possible to their further use in the development of a regulatory document on RPS design.

An improved multi-stage framework for large-scale hierarchical text classification problems using a modified feature hashing and bi-filtering strategy

The classification of large-scale textual dataset is associated with a huge number of instances and millions of features which must be discriminated between large numbers of categories. The task requires the utilization of a defined hierarchy structure and tools that automatically classify instances within the hierarchy known as Large Scale Hierarchical Text Classification (LSHTC). Predicting the labels of instances by the employed classifiers is challenging due to the high number of features. Furthermore, the existing Dimensional Reduction (DR) approaches in cooperation with the LSHTC framework are still quite inefficient. In such a problem, an effective Hierarchical Dimensional Reduction approach can be advantageous in improving the performance of the LSHTC. Therefore, in this paper, we enhance the performance of LSHTC by proposing a Multi-stage Hierarchical Dimensional Reduction (MHDR) approach based on Modified Feature Hashing (MFH) and Hierarchical Bi-Filtering (HBF) method. In addition to alleviating bad collision and result discrepancy, experimental results show that the proposed approach has achieve the best performance in terms of micro-f1 and macro-f1 by recording average scores of 58.47% and 54.77% using TD-SVM, and average scores of 51.14% and 48.70% using TD-LR, respectively. The method also achieved 11% speed-up than the approaches compared.

Использование IS-THE-графов для анализа иерархических структур данных

The article defines the general principles for representing hierarchical structures, such as specializations and categorizations in data models, and class hierarchies in OOP. This representation allows a deeper analysis of the semantics of the subject area and solving the problem of optimal universal implementation of these structures in database systems and object-oriented applications. The possibility for such a generalization is provided by IS-THE relations and IS-THE mappings, on the basis of which IS-THE graphs are constructed. The selection of subgraphs of these graphs according to certain rules generates types of hierarchical structures known in computer science: a single inheritance hierarchy (a specialization hierarchy), a selective inheritance graph (a two-level categorization graph), a two-level multiple inheritance graph.

Цифровой радиоприёмник на основе нейронной сети

One of the areas of digital signal processing used in radio receivers and currently being investigated is the use of artificial neural networks. Signal reception is one of the most difficult theoretical and engineering problems in message transmission. The difficulty lies in the fact that messages need to be extracted from modulated signals, which are exposed to various distorting factors and interference in the radio channel. Therefore, it is desirable to have methods of reception that would be the best (optimal) in the given conditions. There are many different neural network topologies. Single–layer and multi-layer direct propagation are known – perceptrons, recurrent networks, self-organizing networks, as well as hybrid networks (radial-basis, hierarchical classifiers). Each of these types of topologies has its own advantages and disadvantages.The article analyzes current research and development in this area. The implementation of a radio receiver (demodulator) based on a multilayer perceptron is shown and the neural network is trained. Using National Instruments PXI equipment, a study was carried out, which showed that in comparison with an optimal receiver, the probability of a bit error in a receiver based on a neural network is higher, but insignificant. The operation of the receiver under the influence of harmonic interference showed that as the power of the interference increases, the probability of a BER error increases, and the closer the interference is to the carrier frequency, the higher the BER also becomes. Nevertheless, the neural network-based digital receiver under study remains operational at significant levels of interference that are close enough to the carrier frequency of the desired signal.

Security and Privacy Requirements for RFID Schemes in Healthcare: Case Studies, Solutions, and Challenges

RFID technology has proven very useful in many healthcare applications, starting with medical asset tracking, continuing with staff tracking, and reaching medical event and procedure tracking. Security and privacy properties play a fundamental role in this range of applications. The protocols proposed so far highlights many techniques for analyzing these properties, making it difficult to compare the protocols with each other most of the time. In the case of lightweight protocols, the comparisons are even more challenging to achieve because these resort to idealizations that usually have no practical support, such as protocols based on physically unclonable functions (PUFs), or because no security and privacy model is known so far. This paper draws attention to the HPVP model as a possibility for uniformly studying healthcare RFID protocols’ security and privacy properties. We highlight two general patterns that, once encountered in RFID protocols, show that they do not satisfy privacy or mutual authentication requirements. We then illustrate these templates through several proposed RFID protocols for use in healthcare. Then, we propose a Classification of healthcare services based on an extended hierarchy of HPVP privacy classes, and we highlight RFID protocols that reach the respective levels of privacy.

Does the Order Matter? A Random Generative Way to Learn Label Hierarchy for Hierarchical Text Classification

Does the Order Matter? A Random Generative Way to Learn Label Hierarchy for Hierarchical Text Classification

Fast and Effective: Progressive Hierarchical Fusion Classification for Remote Sensing Images

Fast and Effective: Progressive Hierarchical Fusion Classification for Remote Sensing Images

Convolution-Enhanced Bi-Branch Adaptive Transformer With Cross-Task Interaction for Food Category and Ingredient Recognition.

Recently, visual food analysis has received more and more attention in the computer vision community due to its wide application scenarios, e.g., diet nutrition management, smart restaurant, and personalized diet recommendation. Considering that food images are unstructured images with complex and unfixed visual patterns, mining food-related semantic-aware regions is crucial. Furthermore, the ingredients contained in food images are semantically related to each other due to the cooking habits and have significant semantic relationships with food categories under the hierarchical food classification ontology. Therefore, modeling the long-range semantic relationships between ingredients and the categories-ingredients semantic interactions is beneficial for ingredient recognition and food analysis. Taking these factors into consideration, we propose a multi-task learning framework for food category and ingredient recognition. This framework mainly consists of a food-orient Transformer named Convolution-Enhanced Bi-Branch Adaptive Transformer (CBiAFormer) and a multi-task category-ingredient recognition network called Structural Learning and Cross-Task Interaction (SLCI). In order to capture the complex and unfixed fine-grained patterns of food images, we propose a query-aware data-adaptive attention mechanism called Bi-Branch Adaptive Attention (BiA-Attention) in CBiAFormer, which consists of a local fine-grained branch and a global coarse-grained branch to mine local and global semantic-aware regions for different input images through an adaptive candidate key/value sets assignment for each query. Additionally, a convolutional patch embedding module is proposed to extract the fine-grained features which are neglected by Transformers. To fully utilize the ingredient information, we propose SLCI, which consists of cross-layer attention to model the semantic relationships between ingredients and two cross-task interaction modules to mine the semantic interactions between categories and ingredients. Extensive experiments show that our method achieves competitive performance on three mainstream food datasets (ETH Food-101, Vireo Food-172, and ISIA Food-200). Visualization analyses of CBiAFormer and SLCI on two tasks prove the effectiveness of our method. Codes will be released upon publication. Code and models are available at https://github.com/Liuyuxinict/CBiAFormer.

ProFun-SOM: Protein Function Prediction for Specific Ontology Based on Multiple Sequence Alignment Reconstruction.

Protein function prediction is crucial for understanding species evolution, including viral mutations. Gene ontology (GO) is a standardized representation framework for describing protein functions with annotated terms. Each ontology is a specific functional category containing multiple child ontologies, and the relationships of parent and child ontologies create a directed acyclic graph. Protein functions are categorized using GO, which divides them into three main groups: cellular component ontology, molecular function ontology, and biological process ontology. Therefore, the GO annotation of protein is a hierarchical multilabel classification problem. This hierarchical relationship introduces complexities such as mixed ontology problem, leading to performance bottlenecks in existing computational methods due to label dependency and data sparsity. To overcome bottleneck issues brought by mixed ontology problem, we propose ProFun-SOM, an innovative multilabel classifier that utilizes multiple sequence alignments (MSAs) to accurately annotate gene ontologies. ProFun-SOM enhances the initial MSAs through a reconstruction process and integrates them into a deep learning architecture. It then predicts annotations within the cellular component, molecular function, biological process, and mixed ontologies. Our evaluation results on three datasets (CAFA3, SwissProt, and NetGO2) demonstrate that ProFun-SOM surpasses state-of-the-art methods. This study confirmed that utilizing MSAs of proteins can effectively overcome the two main bottlenecks issues, label dependency and data sparsity, thereby alleviating the root problem, mixed ontology. A freely accessible web server is available at http://bliulab.net/ ProFun-SOM/.

ABORDAGEM DA EDUCAÇÃO AMBIENTAL NO ENSINO DE QUÍMICA: UMA ANÁLISE A PARTIR DE ARTIGOS PUBLICADOS NA REVISTA QUÍMICA NOVA NA ESCOLA (QNEsc)

APPROACH TO ENVIRONMENTAL EDUCATION IN CHEMISTRY TEACHING: AN ANALYSIS BASED ON ARTICLES PUBLISHED IN THE JOURNAL QUÍMICA NOVA NA ESCOLA (QNEsc). The knowledge of chemistry connected to environmental problems and sustainable development can provide an opportunity for dynamic, investigative, and directed teaching towards the recognition of chemistry itself as a science and social practice. From this perspective, this work aimed to investigate how discussions involving topics related to environmental education are conducted by researchers and/or professionals in chemical education, based on articles published in the journal Química Nova na Escola (QNEsc). The selection of articles was made considering publications from QNEsc between the years 1995 and 2022, focusing on terms such as environmental education, environment, pollution, environmental aspects, sustainability, and sustainable development present in the title, abstract, and/or keywords of the published articles. The selected texts were analyzed using Iramuteq software, applying correspondence factorial analysis (CFA), hierarchical descendant classification (HDC), similarity analysis, and word cloud using their respective abstracts. The data show that topics such as water, pesticides, waste, and plastic materials (polymers) were the most addressed by the publications, indicating a path that seeks to present and discuss proposals that value the need for sustainable actions, addressing environmental issues that directly impact the way humans use natural resources in the present and future.

Developing a Quality Improvement Implementation Taxonomy for Organizational Employee Wellness Initiatives.

Standardized taxonomies (STs) facilitate knowledge representation and semantic interoperability within health care provision and research. However, a gap exists in capturing knowledge representation to classify, quantify, qualify, and codify the intersection of evidence and quality improvement (QI) implementation. This interprofessional case report leverages a novel semantic and ontological approach to bridge this gap. This report had two objectives. First, it aimed to synthesize implementation barrier and facilitator data from employee wellness QI initiatives across Veteran Affairs health care systems through a semantic and ontological approach. Second, it introduced an original framework of this use-case-based taxonomy on implementation barriers and facilitators within a QI process. We synthesized terms from combined datasets of all-site implementation barriers and facilitators through QI cause-and-effect analysis and qualitative thematic analysis. We developed the Quality Improvement and Implementation Taxonomy (QIIT) classification scheme to categorize synthesized terms and structure. This framework employed a semantic and ontological approach. It was built upon existing terms and models from the QI Plan, Do, Study, Act phases, the Consolidated Framework for Implementation Research domains, and the fishbone cause-and-effect categories. The QIIT followed a hierarchical and relational classification scheme. Its taxonomy was linked to four QI Phases, five Implementing Domains, and six Conceptual Determinants modified by customizable Descriptors and Binary or Likert Attribute Scales. This case report introduces a novel approach to standardize the process and taxonomy to describe evidence translation to QI implementation barriers and facilitators. This classification scheme reduces redundancy and allows semantic agreements on concepts and ontological knowledge representation. Integrating existing taxonomies and models enhances the efficiency of reusing well-developed taxonomies and relationship modeling among constructs. Ultimately, employing STs helps generate comparable and sharable QI evaluations for forecast, leading to sustainable implementation with clinically informed innovative solutions.

A Hierarchical Separation and Classification Network for Dynamic Microexpression Classification

A Hierarchical Separation and Classification Network for Dynamic Microexpression Classification

Deep Learning for Technical Document Classification

In large technology companies, the requirements for managing and organizing technical documents created by engineers and managers have increased dramatically in recent years, which has led to a higher demand for more scalable, accurate, and automated document classification. Prior studies have only focused on processing text for classification, whereas technical documents often contain multimodal information. To leverage multimodal information for document classification to improve the model performance, this paper presents a novel multimodal deep learning architecture, TechDoc, which utilizes three types of information, including natural language texts and descriptive images within documents and the associations among the documents. The architecture synthesizes the convolutional neural network, recurrent neural network, and graph neural network through an integrated training process. We applied the architecture to a large multimodal technical document database and trained the model for classifying documents based on the hierarchical International Patent Classification system. Our results show that TechDoc presents a greater classification accuracy than the unimodal methods and other state-of-the-art benchmarks. The trained model can potentially be scaled to millions of real-world multimodal technical documents, which is useful for data and knowledge management in large technology companies and organizations.

HiCervix: An Extensive Hierarchical Dataset and Benchmark for Cervical Cytology Classification.

Cervical cytology is a critical screening strategy for early detection of pre-cancerous and cancerous cervical lesions. The challenge lies in accurately classifying various cervical cytology cell types. Existing automated cervical cytology methods are primarily trained on databases covering a narrow range of coarse-grained cell types, which fail to provide a comprehensive and detailed performance analysis that accurately represents real-world cytopathology conditions. To overcome these limitations, we introduce HiCervix, the most extensive, multi-center cervical cytology dataset currently available to the public. HiCervix includes 40,229 cervical cells from 4,496 whole slide images, categorized into 29 annotated classes. These classes are organized within a three-level hierarchical tree to capture fine-grained subtype information. To exploit the semantic correlation inherent in this hierarchical tree, we propose HierSwin, a hierarchical vision transformer-based classification network. HierSwin serves as a benchmark for detailed feature learning in both coarse-level and fine-level cervical cancer classification tasks. In our comprehensive experiments, HierSwin demonstrated remarkable performance, achieving 92.08% accuracy for coarse-level classification and 82.93% accuracy averaged across all three levels. When compared to board-certified cytopathologists, HierSwin achieved high classification performance (0.8293 versus 0.7359 averaged accuracy), highlighting its potential for clinical applications. This newly released HiCervix dataset, along with our benchmark HierSwin method, is poised to make a substantial impact on the advancement of deep learning algorithms for rapid cervical cancer screening and greatly improve cancer prevention and patient outcomes in real-world clinical settings.

Structure Feature Extraction for Hierarchical Alarm Flood Classification and Alarm Prediction

Structure Feature Extraction for Hierarchical Alarm Flood Classification and Alarm Prediction

Diagnosis of anemia using non-invasive anemia detector through parametrical analysis

Anemia, characterized by low hemoglobin levels, affects a significant portion of the global population, particularly children and women in low- and lower-middle-income countries. Traditional hematological tests for diagnosing anemia can be invasive, costly, and resource-intensive. This study presents the development of a wearable, non-invasive anemia detector utilizing infrared (IR) technology. The device integrates a NodeMCU microcontroller, an OLED display, a buck converter, an IR transmitter, and a receiver in a compact, wearable package. The IR system measures hemoglobin levels by analyzing the absorption of infrared light emitted through the skin, typically on a fingertip or earlobe. The NodeMCU processes the light absorption data using sophisticated algorithms to determine hemoglobin concentration and facilitates wireless data transfer to smartphones or cloud servers for further analysis. The device is designed with energy efficiency in mind, utilizing a buck converter to ensure optimal power distribution to sensitive components. The OLED display provides users with clear, real-time feedback on their hemoglobin levels, offering a user-friendly interface. Initial testing and calibration with known hemoglobin concentrations demonstrated strong correlations between device readings and traditional lab tests. The device employs a hierarchical ensemble classifier to categorize individuals as healthy, mildly anemic, or severely anemic, aligning with clinical standards. The non-invasive nature of this device, combined with its portability and cost-effectiveness, makes it an ideal solution for point-of-care anemia detection, particularly in resource-limited settings. The integration of IR technology, microcontroller processing, and wireless communication presents a practical tool for healthcare providers, facilitating rapid anemia screening and ongoing monitoring. This innovation holds the potential to improve patient outcomes by enabling early detection and intervention in anemia management.