Weighted Soft Discernibility Matrix with Deep Learning Assisted Face Mask Detection for Smart City Environment

Imène; Afef

doi:https://doi.org/10.54216/IJNS.250116

International Journal of Neutrosophic Science | 2025

Efficient Neutrosophic Optimization for Minimum Cost Flow Problems

In the domain of optimization, linear programming (LP) is recognized as an exceptionally effective method for ensuring the most favorable outcomes. Within the context of LP, the minimum cost flow (MCF) problem is fundamental, with its primary objective being to reduce the transportation costs for a single item moving through a network, under the constraints related to capacity. This network is made up of supply nodes, directed arcs, and demand nodes and each arc has an associated cost and capacity constraint, these factors are certain. However, in practical scenarios, these factors are susceptible to variation due to causal uncertainty. The neutrosophic set theory has surfaced as a challenging approach to tackle the uncertainty that is often encountered in optimization processes. In this manuscript, our primary objective is to address the minimal cost flow (MCF) problem while accounting for the uncertainty inherent in the neutrosophic set. We specifically focus on the cost aspect as SVTN numbers and introduce a new approach based on a customized ranking function handmade for the MCF problem a pioneering endeavor within the field of neutrosophic sets. Additionally, we present numerical example to validate the effectiveness and robustness of our model.  

groups

Shubham Kumar Tripathi mail -

Kottakkaran Sooppy Nisar mail -

Said Broumi mail -

Ranjan Kumar mail

link https://doi.org/10.54216/IJNS.250107

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Automated Credit Card Risk Assessment using Fuzzy Parameterized Neutrosophic Hypersoft Expert Set

In the financial industry, financial fraud is an ever-evolving risk with extreme consequences. Data mining has been instrumental in the recognition of credit card fraud (CCF) during online transactions. CCF recognition, which is a data mining problem, become a challenge owing to its two main reasons - firstly, the profiles of fraudulent and normal behaviors modify continually and then, CCF dataset is extremely lopsided. The implementation of fraud recognition in credit card transactions is tremendously influenced by the sampling methodology on data, detection approach and variable selection utilized. The conception of the neutrosophic hypersoft set (NHSS) is a parameterized family that handles the sub-attributes of the parameter and is an appropriate extension of the NHSS to correctly evaluate the uncertainty, deficiencies, and anxiety in decision-making. In comparison to previous research, NHSS can accommodate additional uncertainty, which is the crucial approach to describe fuzzy datasets in the decision-making algorithm. This study introduces an Automated Credit Card Risk Assessment using Fuzzy Parameterized Neutrosophic Hypersoft Expert Set (ACCRA-FPNHES) technique. In the ACCRA-FPNHES technique, a three-step process is involved. As a primary step, the ACCRA-FPNHES technique designs sparrow search algorithm (SSA) for choosing features. In the second step, the detection of CCF takes place using FPNHES technique. Finally, in the third step, the parameters related to the FPNHES technique can be adjusted by arithmetic optimization algorithm (AOA). The simulation validation of the ACCRA-FPNHES technique can be studied on credit card dataset. The obtained values indicate that the ACCRA-FPNHES technique showcases better performance

groups

Mohammed Abdullah Al-Hagery mail -

Abdalla I. Abdalla Musa mail

link https://doi.org/10.54216/IJNS.250108

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Automated Learning Style Prediction using Weighted Neutrosophic Fuzzy Soft Rough Sets in E-learning Platform

Neutrosophic fuzzy logic (NFL) is a prolongation of classical FL that integrates the neutrosophic conception that handles the indeterminacy concept. This method offers a more comprehensive and flexible architecture to handle inconsistent, uncertain, and indeterminate data, which makes it especially helpful in complicated reasoning and decision-making scenarios where classical FL might be defeated. A learning scheme, which is made from the internet and computer as the main components, is called as an e-learning platform. Although the training might happen on or off campuses, utilizing the internet is an integral part of online learning. In the meantime, to significantly augment the education standard, it is essential to forecast the learning style of the user through supervision and feedback. Nonetheless, it averts the intrinsic relationship amongst e-learning behaviors. There might be technological difficulty ranging from network connectivity issue to users memorizing their username and password while executing and developing an educational program. The learning style prediction in e-learning network is complex one and therefore we recommend a new methodology which employs web mining method for the feature extraction and log files of students from the e-learning network. This study develops an Automated Learning Style Prediction using Weighted Neutrosophic Fuzzy Soft Rough Sets (ALST-WNSFSRS) technique in E-learning Platform. The ALST-WNSFSRS technique mainly aims for the prediction of automated learning styles. Initially, the information is gathered from the Kaggle websites and utilizing a web mining method the feature from the web and log files are pre-processed. The preprocessed information is scrutinized to discover the pattern of approach to learning and later investigated the pattern. Then, the feature patterns are clustered by the fuzzy c-means (FCM) clustering technique and later utilizing the WNSFSRS method, the approach to students learning is anticipated. To improve the performance of the WNSFSRS technique, glowworm swarm optimization (GSO) algorithm is used. The performance of the ALST-WNSFSRS technique is compared with existing models and the results reported the supremacy of the ALST-WNSFSRS technique interms of different measures  

groups

Nasser Nammas Albogami mail

link https://doi.org/10.54216/IJNS.250109

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Robust Diabetic Retinopathy Detection and Grading using Neutrosophic Topological Vector Space on Fundus Imaging

Diabetic retinopathy (DR) is an eye disorder triggered by diabetes that might result in loss of sight. Earlier diagnosis of DR is critical since it might cause loss of sight. Manual diagnoses of DR severity by ophthalmologists are time-consuming and challenging. As a result, there has been considerable attention on designing an automatic technique for DR detection using fundus photographs. In medical science, prognosis and diagnosis are the most challenging tasks due to the presence of fuzziness in medical images and the restricted subjectivity of the experts. Neutrosophic Set (NS) in medical image analysis provides an understanding of the NS concepts, together with knowledge of how to collect, handle, interpret, and analyze clinical images using NS techniques. The neutrosophic set (NS), which is a generality of fuzzy set, provides the overcoming prospect of the restriction of fuzzy-based models for the analysis of medical images. This manuscript develops a Robust Diabetic Retinopathy Detection and Grading using Neutrosophic Topological Vector Space (DRDG-NSTVS) technique on fundus images. The DRDG-NSTVS technique begins with Median Filter (MF) noise removal to optimize the clarity of fundus photographs by successfully eliminating noises. Later, the InceptionV3 is used to perform feature extraction for identifying complicated features and patterns related to DR. The parameter tuning is performed by the moth flame optimization (MFO) technique to ensure superior performance of the model. The final diagnoses and classification of DR are accomplished utilizing the NSTVS classifiers that easily perform the uncertainties inherent in medicinal statistics. The simulation was conducted on a benchmark dataset to examine the proposed model performance. This combined method gives a greatly reliable and accurate solution for the earlier diagnosis and detection of DR

groups

Mohammed Abdullah Al-Hagery mail -

Abdalla I. Abdalla Musa mail

link https://doi.org/10.54216/IJNS.250110

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

On The Topological Spaces of Neutrosophic Real Intervals

In this paper, we present the topological space of intervals based neutrosophic real numbers , where we clarify how neutrosophic real intervals can be expressed according to the neutrosophic partial order relation, and we use these intervals to build a topological space. On the other hand, we use a similar argument to build a topological space over the intervals of refined neutrosophic numbers, with many illustrated and related examples on open and closed sets.    

groups

Raed Hatamleh mail -

Ayman Hazaymeh mail

link https://doi.org/10.54216/IJNS.250111

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

An Outer Generalized Prime System and Some Discrete Examples

Beurling (or generalized) prime system has been defined by Arne Beurling in 1937, and several couthers have been working on this during the last century. This work focuses on addressing some concrete examples of an outer generalized prime system involving Beurling zeta function. The core of this work is to create a discrete generalized prime system under a fixed condition to give a new upper bound for Beurling zeta function.

groups

Ahmed B. AL-Nafee mail -

Faez AL-Maamori mail

link https://doi.org/10.54216/IJNS.250112

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Enhancing Skin Cancer Diagnosis using Cubic Pythagorean Fuzzy Hypersoft Set with Salp Swarm Algorithm

Due to the rapid increase in population density, medical sciences now face a major challenge in the automated detection of diseases. Intelligent system assists health personnel in earlier disease diagnosis and provides reliable treatment to reduce the fatality rates. Skin cancer is one of the most severe and deadliest kinds of cancer. A health professional uses dermoscopic images to manually diagnose skin tumors. This technique can be time-consuming and labor-intensive and needs a considerable level of expertise. The automatic recognition method is essential for the earlier diagnosis of skin tumors. In recent times, N-soft Set model has become widespread, which is a generalization of fuzzy set where all the elements have a membership value in the complement (0 to 1) and in the set (0 or 1). This study presents a Skin Cancer Diagnosis using Cubic Pythagorean Fuzzy Hypersoft Set (SCD-CPFHSS) technique. The presented SCD-CPFHSS technique performs identification of skin cancer using the application of NSs and metaheuristic algorithms. In the SCD-CPFHSS technique, neural architectural search network (NASNet) model derives feature extractors from the dermoscopic image. In addition, the efficacy of the NASNet model can be boosted by the design of salp swarm algorithm (SSA). For skin cancer recognition, the SCD-CPFHSS technique applies CPFHSS model. The experimental outcome of the SCD-CPFHSS methodology was validated using medical dataset. The extensive results pointed out that the SCD-CPFHSS technique reaches better results on skin cancer diagnosis  

groups

Afef Selmi mail -

Imène Issaouı mail

link https://doi.org/10.54216/IJNS.250113

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Blockchain with Single-Valued Neutrosophic Hypersoft Sets Assisted Threat Detection for Secure IoT Assisted Consumer Electronics

The breakthrough technologies of the Internet of Things (IoT) have modernized classical Consumer Electronics (CE) into next-generation CE with high intelligence and connectivity. This connectivity amongst appliances, actuators, sensors, etc., offers automated control in CE and enables better data availability. However, the data traffic has been exponentially increased owing to its decentralization, diversity, and increasing number of CE devices. Furthermore, the static network-based approaches need exclusive management and manual configuration of CE devices.  The generalization of a Neutrosophic Hypersoft Set (NHSS) is a concept of a soft set. This architecture is a mixture of neutrosophic sets with hypersoft sets. Therefore, the study introduce a Blockchain with Single-Valued Neutrosophic Hypersoft Sets Assisted Threat Detection (BCSVNHS-TD) technique for Secure IoT Assisted CE. The presented BCSVNHS-TD technique applies BC technology for secure communication among CEs. For threat detection, the BCSVNHS-TD method introduces the SVNHS model. Also, the parameter selection of the SVNHS method takes place using the chicken swarm optimization (CSO) technique. An extensive set of tests was involved for exhibiting the better effiency of the BCSVNHS-TD method. The experimental results emphasized that the BCSVNHS-TD method reaches optimal results over other techniques  

groups

Mesfer Al Duhayyim mail

link https://doi.org/10.54216/IJNS.250114

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

The Properties of Two-Fold Algebra Based on the n-standard Fuzzy Number Theoretical System

In this paper, we study the two-fold algebra based on the n-standard fuzzy number theoretical system as a special type of two-fold fuzzy algebras, where we study the elementary properties of the algebraic operations defined over this system. Also, we prove many results that describe the relations between two-fold substructures and sub-algebras defined by fuzzy number theoretical systems. On the other hand, we provide many different examples to explain our results.

groups

Raed Hatamleh mail -

Ayman Hazaymeh mail

link https://doi.org/10.54216/IJNS.250115

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new

International Journal of Neutrosophic Science | 2025

Weighted Soft Discernibility Matrix with Deep Learning Assisted Face Mask Detection for Smart City Environment

For smart cities to succeed, substantial developments to take place in roads, city streets, public transportation, houses, businesses, and other aspects of city life must be drawn up. In today’s world, there is a crucial necessity for effective management of cities to reduce the effect of COVID19 disease with increasing population in cities. Multiple metrics had already been taken to lower the infection rate of COVID19, from the beginning of the outbreaks, such as maintaining distance from another person and wearing face masks. Ensuring security in public places of smart cities needs state-of-the-art technology, including computer vision, deep learning and deep transfer learning for automated detection of face masks and monitoring of whether people wear masks accurately.  The achievement of machine learning (ML and) artificial intelligence (AI) techniques in face recognition and object detection makes it fit for the development of FMD methods. The fundamental concept behind the generalized intuitionistic fuzzy soft set is highly productive in making decisions because it considers ways to manipulate an additional intuitionistic fuzzy input from the director to balance any disturbance in the data delivered by the assessment analyst. This manuscript offers the design of Weighted Soft Discernibility Matrix with Deep Learning Assisted Face Mask Detection (WSDMDL-FMD) technique for Smart City Environment. The WSDMDL-FMD technique proficiently discriminates the facial images with the presence or absence of masks. The WSDMDL-FMD technique comprises two stages: Mask RCNN-based face detection and WSDM-based face mask classification. Primarily, the WSDMDL-FMD technique uses Mask RCNN-based face detection. Next, the convolutional neural network (CNN) model derives features from the detected faces and its hyperparameters can be chosen by cuckoo optimization algorithm (COA). For face mask classification, the WSDMDL-FMD technique applies WSDM model. To evaluate the results of the WSDMDL-FMD technique, a series of experiments were involved. The obtained outcomes stated that the WSDMDL-FMD method reaches superior performance than other models  

groups

Imène Issaouı mail -

Afef Selmi mail

link https://doi.org/10.54216/IJNS.250116

Volume & Issue

Vol. Volume 25 / Iss. Issue 1

Details open_in_new