Classification Applications Research Articles

Prognostic values of molecular subtypes and SWI/SNF protein expression in de-differentiated/undifferentiated endometrial carcinoma.

Classification and risk stratification of endometrial carcinoma (EC) has transitioned from histopathological features to molecular classification, e.g. the ProMisE classifier, identifying four prognostic subtypes: POLE mutant (POLEmut) with almost no recurrence or disease-specific death events, mismatch repair deficient (MMRd) and no specific molecular profile (NSMP), with intermediate outcome and p53 abnormal (p53abn) with poor outcomes. However, the applicability of molecular classification is unclear in rare but aggressive histotypes of EC, e.g. de-differentiated and undifferentiated endometrial cancers (DD/UDEC). Here, we aim to assembled a cohort of DD/UDEC from a single institution and analysed the prognostic significance of ProMisE molecular subtypes and the expression of SWItch/sucrose non-fermentable (SWI/SNF) chromatin remodelling complex members, previously implicated in the pathogenesis of DD/UDEC. We accrued 88 DD/UDEC cases, assessed POLE status by Sanger sequencing and performed immunohistochemistry for p53, mismatch repair and SWI/SNF proteins on the tissue microarrays assembled. Assignment of molecular subtypes was possible in 80 tumours; POLE sequencing failed in the remaining eight cases. There were 12 (15%) POLEmut, 44 (55%) MMRd, 14 (17.5%) p53abn and 10 (12.5%) NSMP DD/UDEC. POLEmut DD/UDECs had excellent outcomes, but the other three molecular subtypes all had poor outcomes, with no significant differences among them. The loss of one or more SWI/SNF proteins [AT-rich interactive domain-containing protein 1A (ARID1A), ARID1B, SWI/SNF-related, matrix-associated, actin-dependent regulator of chromatin, subfamily A, member 4 (SMARCA4), SMARCA2], observed in 66% (55 of 83) cases, was not of prognostic significance. These results indicate that all molecular subtypes of DD/UDEC except POLEmut behave in an aggressive fashion. Further study is needed to determine whether these molecular alterations can be targeted with adjuvant therapy, in order to improve outcomes of patients with DD/UDEC.

The Analysis of Conductive Polymer Materials

In recent years, more and more researchers have begun to focus on the study of conductive polymer materials, which not only breaks the traditional cognition that polymer materials are insulators, but also their excellent performance makes them applicable in many aspects. In this paper, a literature review research method will be used to introduce the classification, conducting mechanism, preparation methods and practical applications of conductive polymer materials, based on the research conducted by scientific researchers in recent years. This paper will also combine relevant literature to provide a forecast of their future development. Through research, the paper finds that conductive polymer materials, as a new type of conductive material, have shown great potential in many fields and have broad development prospects in the future. Currently, conductive polymer materials have made considerable contributions in the fields of electronic devices, flexible electronics, sensing technology, and biomedicine. Although conductive polymer materials have made certain progress in many fields, there are still many challenges to be overcome in their development. In the future, with the development of materials science and nanotechnology, conductive polymer materials will achieve breakthroughs in more fields and achieve more widespread applications.

A new approach to construct basic probability assignment and its applications in data classification

In the present work, we have proposed a classification algorithm that uses Dempster Shafer (D-S) evidence theory since it has emerged as an effective tool in handling data classification problems. As basic probability assignment (BPA) is a pre-requisite for applying D-S theory, how to generate it is a hot issue. This paper proposes a novel method for generating basic probability assignments (BPAs) from training data. Dempster Shafer’s (D-S) rule of Combination is utilized for the unification of these BPAs and finally, classify each data item using these unified BPAs. Testing is carried out using some popular benchmark data sets consisting of three classes. Evaluated results show that the classification accuracy is comparatively high.

ARTIFICIAL INTELLIGENCE DRIVEN DRONE OBSERVATION AND PEST CONTROL IN BANANA CROP: A SYSTEMATIC REVIEW

Bananas are the most commonly eaten and significant fruit in global trade. Bananas are produced using a variety of methods and environments. In addition to regularly updating farmers on problems in banana plant leaves, this system aims to discover, diagnose, and treat banana leaf diseases. Customers' wants and lifestyles have changed significantly during the past few decades. These modifications provide additional difficulties for farmers whose output must satisfy consumer needs. Both the farmer and the consumer will benefit from the capacity to categorize agricultural products according to size and quality. In this case, the system gets its input in the form of standard photos of banana leaves taken using various image capture devices. It will then process those photos to identify any diseases and alert the farmer. Additionally, the system will advise the farmer on what to do next, including which fertilizers, herbicides, and agricultural practices to employ in order to prevent illnesses from harming neighboring crops. In this systematic review, useful and efficient methods for identification are presented in works that fall under the categories of image classification, AI/ML, deep learning, and mobile applications.

Vegetable Fibers in Cement Composites: A Bibliometric Analysis, Current Status, and Future Outlooks.

The use of vegetable fibers (VFs) in cement-based composites has increased in recent years owing to their minimal environmental impact and notable particular properties. VFs have aroused interest within the scientific community because of their potential as a sustainable alternative for construction. This study presents a comprehensive bibliometric analysis of VFs in cement composites using data from the Scopus database and scientometric tools to explore publication trends, influential sources, and research directions. Key findings reveal a steady increase in publications, with Construction and Building Materials identified as a leading journal in the field and China and Brazil as prominent contributors in terms of publications and citations. The analysis highlights a strong focus on mechanical properties and durability, reflecting the interest of the scientific community in optimizing VF composites for construction. Furthermore, this study includes a revision of the most influential studies addressing VF classification, durability improvements, and advanced applications of VFs in building applications. Finally, future research opportunities are outlined, emphasizing Life Cycle Assessment (LCA), industry integration, CO2 absorption, and the application of machine learning techniques to advance the development of VF composites. This work provides a comprehensive overview of the field, suggesting future guidelines and promoting collaborative research.

Application of Light-Responsive Nanomaterials in Bone Tissue Engineering.

The application of light-responsive nanomaterials (LRNs) in bone tissue engineering shows broad prospects, especially in promoting bone healing and regeneration. With a deeper understanding of the mechanisms of bone defects and healing disorders, LRNs are receiving increasing attention due to their non-invasive, controllable, and efficient properties. These materials can regulate cellular biological reactions and promote bone cell adhesion, proliferation, and differentiation by absorbing specific wavelengths of light and converting them into physical and chemical signals. In addition, the unique surface morphology and biocompatibility of LRNs enable them to effectively load drugs in bone tissue engineering, achieve precise release, and optimize the bone regeneration process. Through photothermal and photodynamic therapy, these materials also possess antibacterial properties and can play an important role in the repair of infectious bone defects. Although LRNs have shown significant advantages in bone tissue regeneration, a series of challenges still need to be overcome to achieve their widespread and effective clinical applications. This article summarizes the basic principles, classification, and potential applications of LRNs in bone tissue regeneration, aiming to provide reference for future research and clinical applications.

Classification of Strawberry Maturity Stages and Varieties Using Neural Networks Based on Volatile Organic Compounds.

Volatile organic compounds (VOCs) are closely associated with the maturity and variety of strawberries. However, the complexity of VOCs hinders their potential application in strawberry classification. This study developed a novel classification workflow using strawberry VOC profiles and machine learning (ML) models for precise fruit classification. A comprehensive VOC dataset was rapidly collected using gas chromatography-ion mobility spectrometry (GC-IMS) from five strawberry varieties at four maturity stages (n = 300) and visualized through principal component analysis (PCA). Five ML models were developed, including partial least squares discriminant analysis (PLS-DA), decision trees, support vector machines (SVM), Xgboost and neural networks (NN). The accuracy of all models ranged from 90.00% to 98.33%, with the NN model demonstrating the best performance. Specifically, it achieved 96.67% accuracy for single-maturity classification, 98.33% for single-variety classification, and 96.67% for dual maturity and variety classification, along with 98.09% precision, 97.92% recall, and 97.91% F1 score. Feature importance analysis indicated that the NN model exhibited the most balanced reliance on various VOCs, contributing to its optimal performance with the broad-spectrum VOC detection method, GC-IMS. Overall, these findings underscore the potential of NN modeling for accurate and efficient fruit classification based on integrated VOC profiles.

Toward Efficient Credit Card Fraud Detection: Leveraging Quantum Neural Networks and Modified Feature Selection Techniques

Credit cards are a common form of payment not only because they are extremely convenient to use but also because they are widely accepted. Credit cards are not only very easy to use, but they are also readily available. On account of the fact that it is so widely used, there is a substantial amount of concern regarding the protection of sensitive data from fraudulent activities and access by unauthorised individuals. For the purpose of preserving the trust and confidence of users, it is of the utmost importance to make certain that proper security measures are in place. Quantum machine learning (QML) is gaining popularity for classification applications, and a considerable number of the suggestions that have been made for it involve the utilisation of many qubits. This type of learning is becoming increasingly common. It is essential to make every effort to optimise the efficiency and effectiveness of each qubit before adding additional qubits. This should be done before adding more qubits. This is due to the fact that it is probable that these circuits will not always be able to function effectively in the generation of noisy intermediate-scale quantum (NISQ) systems. By utilising a single qubit, the objective of this research is to provide a description of a novel deep quantum neural network that is designed for classification purposes. In comparison to past studies, this network reduces the number of parameters by replicating various tactics that are frequently utilised in convolutional neural networks (CNNs). This is accomplished by reducing the number of parameters. The modified shuffle frog leaping algorithm, also known as MSFLA, is often utilised in order to decide which traits are the most significant while also lowering the amount of computing that is necessary. The purpose is to validate the concept of the first proposal and offer a tested framework for the later development of the application. This will be accomplished through the demonstration of the classification performance of the architecture that is based on a single qubit. Using a dataset that includes records of credit card transactions done by Europeans, the model is assessed in a setting that is reflective of the real world. This is accomplished by using the dataset. A number of components are included in the technique of the proposed model. These components include data pre-processing, feature engineering, ideal selection, evaluation and evaluation, and evaluation and evaluation. The usage of the computational resources provided by Google Colab allows for the training and testing of the model to be carried out with greater efficiency. When compared to individual classifiers, traditional machine learning approaches, and the model that was recommended, it was discovered that the proposed model was more effective in reducing the obstacles connected with detecting credit card fraud. This concluded that the proposed model was more effective. When compared to earlier models, the model that was suggested has a greater degree of performance in terms of accuracy, precision, recall, and F1-score performance characteristics. This is the case when those parameters are measured. The findings that have been provided here provide a foundation for the creation of fraud detection algorithms that are more resilient and flexible. This is something that will become increasingly required as the number of methods that credit card fraud is committed continues to expand.

A Generalized $f$-Divergence With Applications in Pattern Classification

A Generalized $f$-Divergence With Applications in Pattern Classification

Immunoinfiltration Analysis of Mitochondrial Damage-Related Genes in Lung Adenocarcinoma and Construction of a Classification and Prognostic Model Integrated With WGCNA and Machine Learning Algorithms.

Lung adenocarcinoma (LUAD) exhibits molecular heterogeneity, with mitochondrial damage affecting progression. The relationship between mitochondrial damage and immune infiltration, and Weighted Gene Co-expression Network Analysis (WGCNA)-derived biomarkers for LUAD classification and prognosis, remains unexplored. The objective of our research is to identify gene modules closely related to the clinical stages of LUAD using the WGCNA method. Based on the genes within these modules, we constructed machine learning (ML) models for classification and prognosis prediction, thereby facilitating precise diagnosis and personalized treatment of LUAD. Using GeneCards and The Cancer Genome Atlas (TCGA) databases, we screened differentially expressed mitochondrial damage-related genes in LUAD. Immune cell infiltration patterns were assessed using Single-Sample Gene Set Enrichment Analysis (SSGSEA) method. Functional enrichment analyses were conducted to explore biological functions and signaling pathways. Gene modules related to clinical stages of LUAD were identified by WGCNA. ML models were constructed for classification and prognosis prediction, and validated in an independent Gene Expression Omnibus (GEO) dataset. The study revealed a significant relationship between mitochondrial damage and immune infiltration in LUAD. We identified a gene module closely associated with the clinical stages of LUAD. The ML models for classification and prognosis that were constructed demonstrated good effectiveness and generalization capabilities. Mitochondrial damage-related genes are crucial in LUAD progression and linked to immune infiltration. The gene module and models identified have potential applications in LUAD classification and prognosis, offering novel markers for precision medicine. This study uncovers the relationship between mitochondrial damage and immune infiltration in LUAD, paving the way for molecular classification, prognosis prediction, and personalized treatment strategies.

Stimuli‐Responsive Hybrid Microgels for a Sustainable Future: A Comprehensive Review of Classification, Synthesis, and Multifaceted Applications

Stimuli‐Responsive Hybrid Microgels for a Sustainable Future: A Comprehensive Review of Classification, Synthesis, and Multifaceted Applications

Application of Automated Machine Learning and Bagging Techniques to Classify Rice Varieties

We have witnessed increased research investigating digitalisation in the agricultural sector in recent years. In particular, machine learning and artificial intelligence find applications in agricultural product classification, quality control and species identification. The fast-processing times, high accuracy levels and cost-effectiveness offered by digital solutions for quality control and classification accelerate these studies. This study proposes a collaborative learning model utilising Automated Machine Learning and Bagging techniques for rice species detection and classification. The model uses a dataset from the UCI Irvine Machine Learning Repository, which contains characteristics specific to the Osmancık and Cammeo rice varieties grown in Turkey. The data set consists of 3810 data points, 2180 of which belong to Osmancık rice and 1630 to Cammeo rice. During the analysis, MLBox, an Automated Machine Learning library, was used to determine the optimal algorithm (Light Gradient Boosting Machine - LGBM) and its hyperparameters. Later, by applying the Bagging technique within the developed learning model, an accuracy rate of 93.54% was achieved in rice-type classification.

Нейросетевая классификация кандидатов на обучение по их личным и деловым качествам: методология и перспективы применения

Introduction. The The article deals with the application of artificial intelligence technology to enhance the effectiveness of professional psychological selection within internal affairs bodies. The importance of this study lies in the need to carefully select and train candidates for service in order to form the backbone of the personnel of the internal affairs bodies in the Russian Federation. In order to select the right candidates, it is necessary to enhance the process of differential ability. Additionally, this study highlights the scientific need to improve psychological support for personnel, development of methodology, methods and procedures of psychological diagnostics as a field of scientific knowledge. The research methods are general scientific methods of process information to organise knowledge about professional psychological selection and the application of neural network classification. The empirical methods used are the construction of a decision tree and neural network classification. The research sample comprises candidates for training in the educational organisation of internal affairs bodies. The results of the research can be summarised as follows: a) decision rules (decision tree) for the classification of applicants, revealing the regularities of their distribution into categories of professional suitability based on their personal and professional qualities; b) a neural network model for the differentiation of applicants into categories of professional suitability based on their personal and professional qualities. This model has shown good accuracy in selecting candidates. Theoretical results are findings about the possibility of individualisation of psychological support and improving educational work with students who have been referred to different categories of professional suitability on the basis of individual combinations of personal and professional qualities. Theoretical results also concern the improvement of educational work with students. The potential for enhancing the capacity of psychodiagnostic techniques to differentiate between individuals and the procedure for examining the personal and professional qualities of candidates for service, which reduce the risk of inaccuracies in their measurement, are outlined. The conclusion outlines the theoretical and methodological prospects for applying neural network classification method as a component of artificial intelligence technology in the selection of candidates for service in the internal affairs bodies of the Russian Federation.

Comprehensive Review on Plant Disease Detection Techniques

Plant diseases present a significant challenge to global agricultural productivity and food security. Traditional methods for identifying plant diseases rely heavily on manual observation and expertise, which are prone to error and scalability issues. Recent advancements in artificial intelligence, particularly machine learning (ML) and deep learning (DL), have revolutionized plant disease detection. This paper reviews eight key research studies exploring these advancements, focusing on techniques such as Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and hybrid models that incorporate preprocessing and feature extraction. While existing methods achieve impressive accuracy in controlled conditions, their application to real-world environments remains limited by dataset diversity, environmental variability, and computational demands. This paper also highlights the importance of developing lightweight, scalable, and adaptable models. Future recommendations include expanding datasets, integrating Internet of Things (IoT) technologies, and exploring multimodal approaches for holistic plant health monitoring. Plant diseases significantly impact global food security, causing severe economic and productivity losses. This review consolidates findings from eight key research studies on plant disease detection using artificial intelligence, specifically machine learning and deep learning approaches. Techniques such as Convolutional Neural Networks (CNNs), Support Vector Machines (SVMs), and K-Nearest Neighbours (KNN) are evaluated, with a focus on their application in image classification and preprocessing. Challenges such as dataset diversity, environmental variability, and computational complexity are analysed. Future recommendations include developing scalable models, creating diverse datasets, and integrating real-time systems like drones and IoT technologies to revolutionize agricultural practices. Keywords: Plant Disease Detection, Machine Learning, Deep Learning, Image Processing, PlantVillage Dataset, IoT, Agricultural Technology , Convolutional Neural Networks , Support Vector Machines , K-Nearest Neighbours , computational complexity , multimodal approaches .

Digital Characteristics of Spatial Layout in Urban Park Scene Space: Spatial Classification, Quantitative Indicators, and Design Applications Based on Completed Park Cases

Urban parks are important components of the urban green space system, providing residents with a variety of leisure options. The design’s focal point is the spatial layout of the different use scenarios within these parks. Previous studies have largely concentrated on the scene and macro layout of urban public spaces but have not thoroughly investigated the digital characteristics and corresponding parametric methods for the layout of different internal use scenes in urban green spaces. This research selected 18 urban parks from various global regions as case samples and categorized eight typical park scene space types based on common activity scene requirements in park design using AutoCAD to identify vector boundaries in each sample. To examine the digital characteristics of these scene space types, a quantitative index system was established, including spatial density and scale indicators, spatial connection relationship indicators, and spatial unit morphology indicators. The analysis of these indicators across the samples shows that the number of scene spaces is positively correlated with the total park area, while natural experience scenes constitute the largest proportion of urban parks. Different scene types exhibit distinct spatial layouts; for example, circulation spaces demonstrate high connectivity due to their role in directing visitor flow. Some spaces exhibit a more fixed scale and size, while others vary considerably. Finally, this research develops a parametric design framework using the Grasshopper platform. By taking a park in Nanjing as a case study, this paper illustrates how to utilize digital layout features to generate scene space layouts, offering insights into intelligent generative design. This approach provides a structured method to enhance urban park design through the application of digital and parametric tools, contributing to the broader field of urban park design.

Machine learning for brain tumor classification: evaluating feature extraction and algorithm efficiency

Uncontrolled fast cell growth causes brain tumors, posing a significant threat to global health and leading to millions of deaths annually. Early cancer detection is crucial to save lives. The purpose of this study is to investigates the capability of machine learning algorithms and feature extraction methods to detection and classification of brain tumors. We implemented six machine learning algorithms and three features extraction methods, including Image Loading, HOG, and LBP. The objective of this study is to identify best combination of machine learning and features extraction method for brain tumor detection and classification. This study utilized two Brain Tumor MRI Datasets downloaded from Kaggle. Our analysis revealed that Random Forest emerged as the most effective classifier by achieving an accuracy of 99% with image loading feature extraction method based on different metrics, closely followed by SVM and Logistic Regression. However, performance varied with KNN, Naive Bayes, and Decision Tree, highlighting the importance of tailored approaches for optimal classification accuracy. Further optimization and experimentation are crucial for improving algorithm performance in real-world applications of brain tumor classification. A case study with interpretable machine learning is also presented in the paper.

Variability in hot sub-luminous stars and binaries: Machine-learning analysis of Gaia DR3 multi-epoch photometry

Hot sub-luminous stars represent a population of stripped and evolved red giants that is located on the extreme horizontal branch. Since they exhibit a wide range of variability due to pulsations or binary interactions, it is crucial to unveil their intrinsic and extrinsic variability to understand the physical processes of their formation. In the Hertzsprung-Russell diagram, they overlap with interacting binaries such as cataclysmic variables (CVs). By leveraging the most recent clustering algorithm tools, we investigate the variability of 1,576 candidate hot subdwarf variables using comprehensive data from Gaia DR3 multi-epoch photometry and Transiting Exoplanet Survey Satellite (TESS) observations. We present a novel approach that uses the t-distributed stochastic neighbour embedding and the uniform manifold approximation and projection dimensionality reduction algorithms to facilitate the identification and classification of different populations of variable hot subdwarfs and CVs in a large dataset. In addition to the publicly available Gaia time-series statistics table, we adopted additional statistical features that enhanced the performance of the algorithms. The clustering results led to the identification of 85 new hot subdwarf variables based on Gaia and TESS light curves and of 108 new variables based on Gaia light curves alone, including reflection-effect systems, HW Vir, ellipsoidal variables, and high-amplitude pulsating variables. A significant number of known CVs (140) distinctively cluster in the 2D feature space among an additional 152 objects that we consider candidates for new CVs. This study paves the way for more efficient and comprehensive analyses of stellar variability from ground- and space-based observations, and for the application of machine-learning classifications of candidate variable stars in large surveys.

Unlocking the Power of Antioxidant Nanoparticles: Insights into Classification, Formulation, Characterization, and Biomedical Applications

Unlocking the Power of Antioxidant Nanoparticles: Insights into Classification, Formulation, Characterization, and Biomedical Applications

Pengembangan Aplikasi Klasifikasi Gambar Foto Kegiatan ‘Remaja Bagi Kristus’ Gereja Isa Almasih Kega

This study aims to develop an image classification application based on a website. An application will be made to be able to classify pictures of activities at Youth For Christ, KEGA Isa Almasih Church. In this study using the CNN (Convolutional Neural Network) method to create a model. The machine learning model used in building this application uses the Tensorflow software library. The software development method used is the incremental method. The dataset used in this study was collected as much as 600 image data and will be divided into 3 data variables, namely praying, preaching, and worship. The result of the application that is built is a simple website-based image classification application so that users can easily use the application. By utilizing the HTML programming language in making application interfaces and Python as a programming language in creating machine learning models used in applications, in making this application the researcher also uses Flask as a backend for the application.

A Survey on Deep Learning for Few-Shot PolSAR Image Classification

Few-shot classification of polarimetric synthetic aperture radar (PolSAR) images is a challenging task due to the scarcity of labeled data and the complex scattering properties of PolSAR data. Traditional deep learning models often suffer from overfitting and catastrophic forgetting in such settings. Recent advancements have explored innovative approaches, including data augmentation, transfer learning, meta-learning, and multimodal fusion, to address these limitations. Data augmentation methods enhance the diversity of training samples, with advanced techniques like generative adversarial networks (GANs) generating realistic synthetic data that reflect PolSAR’s polarimetric characteristics. Transfer learning leverages pre-trained models and domain adaptation techniques to improve classification across diverse conditions with minimal labeled samples. Meta-learning enhances model adaptability by learning generalizable representations from limited data. Multimodal methods integrate complementary data sources, such as optical imagery, to enrich feature representation. This survey provides a comprehensive review of these strategies, focusing on their advantages, limitations, and potential applications in PolSAR classification. We also identify key trends, such as the increasing role of hybrid models combining multiple paradigms and the growing emphasis on explainability and domain-specific customization. By synthesizing SOTA approaches, this survey offers insights into future directions for advancing few-shot PolSAR classification.