Detecting Cyberbullying and Hate Speech in Regional Languages Using Hybrid Deep Learning and NLP Models

Ganesh; Kumarganesh; Elayaraja; Thiyaneswaran

doi:https://doi.org/10.54216/JCHCI.090206

International Journal of Neutrosophic Science | 2025

Unconstrained Neutrosophic Nonlinear Programming Problems Gradient Projection Method

Nonlinear programming is one of the most important methods used to obtain the optimal solution to many real-world problems. Given the importance of this method, numerous studies and research have been conducted in recent years with the aim of providing methods that help find the optimal solution. These studies and research have resulted in a basic structure used to find these solutions. This structure initially indicates that the optimal solution can be found at any boundary point in the feasible region, at a point within the feasible region, or at a discontinuity point. In this research, we present some of the important foundations and principles of nonlinear programming and the gradient projection method used in searching for the optimal solution to unrestricted nonlinear programming problems. We will reformulate these foundations and principles using neutrosophic logic concepts as a complement to our previous research, the aim of which is to provide a new vision for some operations research methods, a neutrosophic vision. Our focus will be on the improvement these concepts offer when used in the field of applied mathematics, through the more accurate and comprehensive solutions we obtain, which provide a margin of freedom commensurate with the Given the reality we live in, and the changes that can occur to the data of the actual issue under study, this requires decision makers to prepare many appropriate alternatives for each change.

groups

Maissam Jdid mail -

Florentin Smarandache mail

link https://doi.org/10.54216/IJNS.260320

Volume & Issue

Vol. Volume 26 / Iss. Issue 3

Details open_in_new

International Journal of Neutrosophic Science | 2025

Developed acceptance sampling plans for the Shanker distribution based on truncated life tests

This paper introduces new acceptance sampling plans for situations where the life test is terminated at a predetermined time. The minimum sample sizes needed to guarantee a specified average lifetime are determined for different acceptance numbers, confidence levels, and ratios of the fixed test duration to the defined average lifetime. The Shanker distribution is adopted to represent the lifetimes of test units, with its mean serving as the quality indicator. Furthermore, the operating characteristic function values for the proposed sampling plans, along with the associated producer's risk, are provided. Examples are included to demonstrate how to use the tables effectively. An application of a real data set is used to illustrate the usefulness of the suggested acceptance sampling plans.

groups

Amer Ibrahim Al-Omari mail -

Rehab Alsultan mail

link https://doi.org/10.54216/IJNS.260321

Volume & Issue

Vol. Volume 26 / Iss. Issue 3

Details open_in_new

International Journal of Neutrosophic Science | 2025

Crossing Cubic Structures Applied to Hoop Algebras

Recent years have witnessed remarkable developments in fuzzy logic, with interval-valued fuzziness and negative structures emerging as powerful tools for modeling inaccurate phenomena. The crossing cubic structures (CCs), as a generalization of the bipolar fuzziness structures, represent a comprehensive mathematical framework capable of dealing with a wide range of fuzziness and contradictory data, thus expanding research prospects in this area. This paper has made a new contribution to some algebraic structures by investigating the concept of CCs on algebraic substructures in a hoop algebra. The concepts of crossing cubic sub-hoops (CC − SHs) and crossing cubic filters (CCFs) are introduced, and a deeper understanding is sought to analyze their characteristics. The effect on the relationship between CC − SHs and CCFs is revealed, and the characterizations of CC − SHs and CCFs are analyzed.

groups

Anas Al-Masarwah mail -

Fawziah Alharthi mail -

Noor Bani Abd Al-Rahman mail

link https://doi.org/10.54216/IJNS.260322

Volume & Issue

Vol. Volume 26 / Iss. Issue 3

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

PhyGital Fit: An AI-Driven Virtual Footwear Solution Integrating Generative AI, AR and Foot Morphology Analysis for Personalized Fit

Rapid development has been seen in Artificial Intel license (AI), which has transformed the retail industry, including online shopping. Selecting the right size of shoes that varies with brands and design is one of the biggest challenges in the E-Commerce footwear industry. This research focuses on AI Powered virtual shoe fitting system using Lens Studio Software. In this, customers are able to try shoes virtually through augmented reality and customized 3D foot models. This innovation solves size issues and benefits online footwear retailers, resulting in greater customer satisfaction. The role of Lens Studio software includes the creation of customized shoes, 3D shoes models, lenses, and size accuracy with the foot tracking mechanism.

groups

Abhimanyu Sangale mail -

Nikita Bhawar mail -

Rutuja Gholap mail -

Bhoomi Raut mail -

Kanchan Suryavanshi mail

link https://doi.org/10.54216/JCHCI.090201

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

Climate Change and Sustainability: A Review

Climate change, driven by human activities like burning fossil fuels, deforestation, and industrial agriculture, is one of the most urgent global challenges. The rise in greenhouse gases (GHGs), such as carbon dioxide (CO₂), methane (CH₄), and nitrous oxide (N₂O), is contributing to global warming, sea level rise, and extreme weather events, with developing nations being particularly vulnerable. To address this, sustainability has become a key focus, involving the need to meet present demands without compromising the ability of future generations to meet theirs. Mitigation strategies include reducing emissions, transitioning to renewable energy sources like solar, wind, and hydropower, improving energy efficiency, and using reforestation to absorb carbon dioxide. Adaptation efforts, such as drought-resistant crops and resilient infrastructure, help communities cope with the impacts of climate change. The circular economy, which emphasizes resource efficiency, waste reduction, and recycling, further supports environmental sustainability. Governments, corporations, and individuals must also prioritize social justice, ensuring that underserved areas most affected by climate change receive the necessary support. Through collective action, we can work towards a sustainable future for all.

groups

Balaji Vijayan Venkateswarulu mail -

Ankitha K. mail -

Chandana L. mail -

Likitha M. mail

link https://doi.org/10.54216/JCHCI.090202

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

Proposed Strategies for Sustainable Agriculture Domain Chatbot with Blockchain Development

The farm sector is challenged by various factors, such as climate volatility, ineffective resource management, and data security. In this paper, a new methodology is proposed where blockchain technology is combined with a chatbot platform to offer farmers real-time, secure, and accurate crop suggestions. Blockchain allows data integrity to be guaranteed, reducing risks from data tampering. The chatbot is an interactive platform, where farmers can enter soil parameters, location, and weather. The system processes these inputs and gives optimal crop recommendations based on past data and predictive analytics. The proposed solution is enhancing sustainable agriculture practices, boosting productivity, and ensuring long-term food security.

groups

S. Rahini Sudha mail -

Uma mageshwari D. mail -

Vaishnavi N. G. mail -

Sumetha S. S. mail -

Subbalakshmi B. V. mail -

Poovizhi A. mail

link https://doi.org/10.54216/JCHCI.090203

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

Review of Machine Learning Technique based Prediction Model for Phishing Websites Detection

Phishing attacks have emerged as a significant cybersecurity challenge, targeting individuals and organizations by tricking users into revealing sensitive information through deceptive websites. Traditional phishing detection methods, such as blacklists and heuristic-based approaches, struggle to keep pace with the rapid evolution of phishing techniques. Machine learning-based predictive models offer a promising solution by analyzing website attributes, URL structures, and behavioral patterns to distinguish between legitimate and phishing websites. This paper provides a comprehensive review of various machine learning techniques, including decision trees, support vector machines (SVM), random forests, deep learning models, and ensemble methods, employed in phishing website detection. It explores feature selection strategies, dataset characteristics, performance evaluation metrics, and real-world implementation challenges. Furthermore, the study discusses recent advancements such as adversarial resilience, natural language processing (NLP) integration, and real-time phishing detection frameworks. The review highlights existing research gaps and future directions to enhance phishing detection accuracy, scalability, and adaptability in evolving cybersecurity landscapes.

groups

RishiKesh Dube mail -

Twinkle Sharma mail -

Damodar Tiwari mail -

Kailash Patidar mail

link https://doi.org/10.54216/JCHCI.090204

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

Injury prediction and prevention for cricket players using AI

Cricket is a physically demanding sport that exposes players to various acute and chronic injuries. Preventing these injuries is crucial for maintaining peak performance and prolonging careers. This project leverages artificial intelligence (AI) and machine learning (ML) to analyze key player data, including biomechanics, workload, fatigue, and mental stress, to assess and mitigate injury risks. Wearable sensors and tracking systems continuously monitor player movements, workload, and physiological parameters, providing real-time insights into their physical condition. By detecting patterns that indicate potential injury risks, the AI model enables early intervention through personalized training modifications and recovery programs. This proactive approach minimizes injuries, optimizes player fitness, and enhances performance. Ultimately, integrating AI-driven injury prevention strategies in cricket ensures better player management, increased longevity, and improved overall team efficiency.

groups

A. V. Adlin Grace mail -

Sanjay Kumar S. mail -

Rajesh S. mail -

Ragul Doss R. mail

link https://doi.org/10.54216/JCHCI.090205

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new

Fusion: Practice and Applications | 2025

Hybrid Deep Learning Models for Finger Vein Biometric Authentication with Experimental Insights and Robust Performance Evaluation

The proposed method creates an advanced Deep Residual Convolutional Neural Network (DR-CNN) for finger vein pattern recognition to enhance both accuracy and computational efficiency of the system. The framework implements DR-CNN to handle the reduction of dimensions together with feature extraction while resolving traditional CNN models' overfitting issues. This research utilizes 6,000 images from the VERA and PLUSVein FV3 and MMCBNU_6000 and UTFV databases which form 80% training data and 20% testing data. The ImageNet training includes 4 pooling layers while also using 4 fully connected layers as well as 13 convolutional layers. The DR-CNN classifier achieves optimal authentication-performance through its implementation of Gray Level Co-occurrence Matrices (GLCM) and Scale-Invariant Feature Transform (SIFT) for extracting features. A performance assessment based on accuracy, sensitivity, specificity, F1-score, false acceptance rate (FAR) and false rejection rate (FRR) proves that DR-CNN surpasses traditional techniques. With its implementation of 5,000 images the proposed model demonstrates better accuracy (94.39%) than CNN (92.45%), RNN (88.99%) and DNN (85.91%). Tests show that the system processes 25,000 images within 2.43 milliseconds establishing fast computation speeds. DR-CNN achieves robustness through minimum mean absolute error values of 19.34. The proposed DR-CNN model delivers a 97.8% recognition rate together with a 0.83% error rate which proves its effectiveness for biometric security applications.

groups

Hashem Alyami mail

link https://doi.org/10.54216/FPA.200108

Volume & Issue

Vol. Volume 20 / Iss. Issue 1

Details open_in_new

Journal of Cognitive Human-Computer Interaction | 2025

Detecting Cyberbullying and Hate Speech in Regional Languages Using Hybrid Deep Learning and NLP Models

The rise of social media platforms has led to an increase in cyberbullying and hate speech, which can have severe consequences on individuals and communities. The detection of harmful content, especially in regional languages, remains a significant challenge due to the linguistic diversity, informal expressions, and limited datasets available for training machine learning models. This paper proposes a hybrid deep learning and natural language processing (NLP) model for the detection of cyberbullying and hate speech in regional languages. The model combines convolutional neural networks (CNNs) and recurrent neural networks (RNNs) with advanced NLP techniques such as sentiment analysis and context-aware feature extraction. Preliminary experiments show that the proposed model achieves an accuracy of 86.7% for hate speech detection and 82.3% for cyberbullying detection in regional language datasets. Furthermore, the hybrid model outperforms traditional machine learning techniques by 15% in terms of precision and recall. This approach demonstrates the potential of combining deep learning and NLP to address the challenges of detecting harmful content in diverse linguistic environments.

groups

Ganesh C. mail -

Kumarganesh S. mail -

Elayaraja P. mail -

Thiyaneswaran B. mail

link https://doi.org/10.54216/JCHCI.090206

Volume & Issue

Vol. Volume 9 / Iss. Issue 2

Details open_in_new