Investigative Study of Fuzzy Logic Applications in Natural Disaster Management: A Comprehensive Review

Wed, 01 Oct 2025 00:00:00 +0000

Climate change, together with urban development and environmental destruction, results in more frequent and severe natural disasters, which include earthquakes, floods, hurricanes, and wildfires. Disaster management systems experience major difficulties when uncertainty and incomplete information and changing conditions are present during these events. The lack of proper modelling of disaster data uncertainty within traditional mathematical approaches results in suboptimal decisions for prediction and preparation and response phases. Fuzzy logic operates as a soft computing technique which effectively handles imprecise data through linguistic variables combined with membership functions and rule-based inference systems. The review examines the entire disaster management process through fuzzy logic applications which span mitigation and preparation stages and extend to response operations and recovery phases. The paper provides an organized overview of applications based on disaster types and fuzzy logic variants and methods while analysing performance advantages together with limitations and integrating new technologies such as IoT and big data and machine learning. This research aggregates recent studies to reveal essential directions and unexplored areas and forthcoming developments which aim to support both research and practical development of resilient smart disaster management systems.

Machine Learning-Driven Evolutionary Fuzzy Clustering for High-Dimensional Genomic Data Analysis

Fri, 28 Nov 2025 00:00:00 +0000

Clustering and classification Genomic data analysis is a fundamental part of contemporary bioinformatics and precision medicine, yet clustering and classification of high-dimensional genomic data is extremely challenging due to noise and uncertainty. The traditional methods of clustering (such as hard partitioning) techniques are not practical in biological data that display overlapping cluster structures, and existing fuzzy clustering methods (such as Fuzzy C-Means (FCM)) are sensitive to initialisation and are likely to converge on local optima. This paper presents a high-dimensional genomic-specific Machine Learning-based Evolutionary Fuzzy Clustering (MLEFC) architecture to ensure these challenges are addressed. The suggested framework makes the power of the evolutionary algorithms such as Genetic Algorithms (GA) and Particle Swarm Optimization (PSO) to maximize centres of fuzzy clusters and membership degrees to address the weakness of traditional clustering. Also, machine learning models are incorporated on cluster validation, predictive analysis, and enhanced biological interpretability. This framework is tested with benchmark genomic data, and results show that it has significantly better clustering accuracy, stability and scale than the traditional ones. Additionally, the biological validation underscores the potential of the framework in finding functional groups of genes, disease subtypes and candidate biomarkers. MLEFC introduces a powerful and interpretive tool to genomic data mining by incorporating fuzzy logic, evolutionary computing, and machine learning and has direct applications to biomarker discovery, cancer subtype classification, and personalized medicine.

Journal of Fuzzy Sets and Fuzzy Logic Design (e-ISSN: 3049-0227)

Investigative Study of Fuzzy Logic Applications in Natural Disaster Management: A Comprehensive Review

Machine Learning-Driven Evolutionary Fuzzy Clustering for High-Dimensional Genomic Data Analysis