Climate change is intensifying environmental pollution, altering both pollutant distribution and the effectiveness of biological remediation strategies. Predicting pollution trends and designing adaptive remediation approaches are critical for sustainable ecosystem management. Traditional modeling techniques often struggle with the non-linear, multi-factorial nature of environmental systems. There is a pressing need for robust computational models that can accurately forecast pollution dynamics while optimizing biological remediation strategies under uncertain climate scenarios. Existing methods frequently face challenges in convergence speed, local optima avoidance, and adaptability to complex environmental datasets. This study introduces a Levy flight-enhanced soft computing framework, integrating recent meta-heuristic algorithms with fuzzy logic and neural computation. The approach leverages Levy flight-inspired exploration to improve global search capabilities, enabling better parameter tuning and predictive accuracy. Historical pollution datasets, climatic variables, and biological remediation performance indicators were used to train and validate the model. The framework evaluates the influence of temperature fluctuations, precipitation patterns, and pollutant load on remediation efficiency, providing actionable insights for environmental management. Experimental results demonstrate that the proposed Levy-based soft computing model achieves superior predictive accuracy, with a 15–20% improvement over conventional heuristic approaches in forecasting pollutant concentrations. Additionally, the framework identifies optimal biological remediation strategies, enhancing contaminant removal efficiency by up to 18% under varying climate scenarios. Sensitivity analysis highlights key climatic factors influencing remediation performance, confirming the model’s robustness and adaptability to dynamic environmental conditions.

Depression affects individuals globally, necessitating efficient diagnostic tools. This study introduces an advanced unsupervised hybrid approach, that automatically converts binary-labelled depression datasets into multi-class datasets by integrating a rule-based system with Large Language Models (LLMs). The rule-based system employs the Beck Depression Inventory-II tool that can be used to classify depression levels based on predefined scoring rules, and these rules are segregated into clusters based on score ranges from 0-3. LLMs employ fine-tuned large language Model Meta AI2 (LLaMa2) to generate domain-specific embedding from social media posts. By harnessing LLMs’ contextual understanding, both BDI rules and social media posts are embedded, thereafter cosine similarity is applied to calculate semantic similarities. Based on the similarity score, each post is assigned to the most similar BDI cluster, with the highest similarity score, creating a refined multiclass depression cluster. To evaluate clustering effectiveness, the silhouette score was computed, yielding an average score of 0.45, indicating moderate clustering quality. Additionally, 30% of the binary depression dataset was manually labelled by clinical experts. The Normalized Mutual Information (NMI) score of 0.53 further validated the method, showing strong alignment between the generated clusters and expert-labelled data. This approach enhances depression severity classification, providing a scalable, efficient, and accurate tool for researchers and practitioners.

The usage of social media to interchange ideas and facts has increased exponentially due to technological advancements. Platforms for video sharing, like YouTube, have distinctive environments and architecture that people use for entertainment, education, and to keep themselves updated. YouTube is one of the most frequently used social media platforms, and users can connect to it by viewing, sharing opinions through comments, liking and disliking videos. A viewpoint or judgement formed about anything is referred to as an opinion. It can be collected and used to check knowledge, suggest the author with new video ideas, and analyze user behaviour. In this study, the data extracted from the free video-sharing platform YouTube concerning the ‘Air India Flight Urination Case’ was observed recently to recognize people’s opinions on national and international levels. Based on approximately 10,000 comments about the incident, models are applied to classify and investigate the sentiments. This investigation uses TF-IDF and Bag of Words (BoW) text modelling techniques and observed that BoW performs better than TF-IDF. Moreover, Naive Bayes, Logistic Regression, Decision Tree, Support Vector Machines, and some ensemble algorithms like Random Forests, Gradient Boost, and Voting Classifier combining (Support Vector Machine, Decision Tree, Logistic Regression and Random Forest) with soft and hard voting had been applied and found that Support Vector Machine has the highest classification accuracy of 84%.

In the digital age, password security is an important part of cybersecurity, especially for university students who often access online platforms for academic and personal use. This study examines the perceptions and perceptions of password security under university students, and examines knowledge, practices, and attitudes to protect online accounts from Botswana. Research uses a survey based on a survey to assess the understanding of students strongly creating passwords, using multi-factor verification and their vulnerability to cyber threats such as phishing and theft of credentials. The findings reveal a significant abyss between consciousness and implementation, with many students acknowledging the importance of password security, but do not take robust protective measures. Furthermore, in this study, Password security practices are highlighted, including comfort, lack of cybersecurity and low risk of cyberattacks. This research emphasizes the need for an increased campaign for education and awareness of cyber security within academic institutions to bridge the abyss between knowledge and practice. The study recommends integrating password security training in the curriculum, to promote the importance of password management and encourage the students to follow the rules and procedures for regular passwords change and multi-factor verification. By strengthening the awareness of password security with students is essential to improve cybernetic risks and ensure the protection of personal and institutional data.

Kidney stone diagnosis is one of the sensitive issues in personal healthcare. Detecting kidney stones early can play a vital role in avoiding chronic kidney diseases and related surgical procedures. However, due to several associated issues, identifying a kidney stone in the early stages can be very difficult. In this research, a classification model for automated diagnosis of kidney stones utilizing coronal computed tomography (CT) images is suggested. Due to low resolution and the presence of noise, every image is passed through an image enhancement step before feeding into a VGG-19 based CNN Model. The training dataset used contains 1799 cross-sectional CT scan images from 433 individuals. Data augmentation is carried out to avoid overfitting of the deep model. The developed model can correctly identify kidney stones of even tiny size with a 97.62% precision, 98.79% recall, and 98.62% accuracy. The developed model performs better than recent similar work and is suitable for e-healthcare systems. It demonstrates that such deep-learning-based techniques can be utilized to solve other similar issues in urology.

Rapid urbanization has intensified air pollution and climate-related risks, challenging sustainable city planning. Green infrastructure (GI) has emerged as a vital strategy for mitigating these environmental stressors. However, selecting optimal GI interventions requires the integration of multiple, often conflicting, criteria such as pollution reduction efficiency, cost-effectiveness, and resilience to climate variability. Conventional decision-making approaches in urban planning often struggle to handle the uncertainty and complexity inherent in multi-criteria environmental assessments. This limitation hampers the identification of effective GI solutions tailored to specific urban contexts, leading to suboptimal pollution and climate risk mitigation. This study proposes a Deep Neural Fuzzy Multi-Criteria Decision Support System (DNF-MCDSS) for prioritizing GI strategies in urban environments. The framework combines fuzzy logic with deep neural networks to model uncertainty and non-linear relationships among environmental, social, and economic criteria. Input data covering air quality indices, climatic variables, land use patterns, and socioeconomic factors are processed through the hybrid network to generate a ranked list of GI interventions. The model’s performance is evaluated using case studies from metropolitan regions, with validation against expert assessments and conventional multi-criteria decision methods. Experimental results demonstrate that the proposed DNF-MCDSS consistently identifies GI strategies that maximize pollution reduction while enhancing climate resilience. For example, green roofs, urban tree corridors, and constructed wetlands were prioritized in scenarios with high pollution loads and extreme heat events. Compared with traditional weighted-sum and AHP methods, the framework achieved a 15–20% improvement in alignment with expert recommendations, showing its ability to capture complex interdependencies and uncertainty in urban environmental planning.

Cyberattack detection in industrial control systems (ICS) is critical to ensuring the security and resilience of essential infrastructure, such as water treatment and distribution networks. However, existing anomaly detection methods often struggle with capturing complex temporal dependencies and differentiating between cyberattacks and normal operational variations. In this study, we propose a novel Transformer- based approach with hybrid positional embeddings for detecting cyber- attacks in multivariate time series data. Our method integrates learnable, sinusoidal, and rotary position embeddings, enabling the model to effectively capture both absolute and relative temporal relationships. This hybrid embedding strategy addresses key limitations of conventional Transformers in handling time-series data by improving the encoding of temporal dependencies. We evaluate our approach on two widely used cybersecurity datasets: Secure Water Treatment (SWaT) and Water Distribution (WADI), which simulate real-world industrial cyber-physical system (CPS) attacks. Our model outperforms state-of-the-art baselines, achieving high detection accuracy and robust anomaly identification. Additionally, an ablation study demonstrates the contribution of hybrid positional embeddings in improving cyberattack detection performance. This work enhances AI driven security frameworks for industrial systems by providing a scalable and effective solution for cyber threat monitoring in critical infrastructures.

Metaheuristics have been used to solve combinatorial optimization problems in recent decades. Metaheuristics inspired by various natural phenomena have been proposed due to their optimization characteristics. The Imperialist Competitive Algorithm (ICA) is one such metaheuristic inspired by the socio-political process of imperialism. ICA has become popular due to its extensive applications in various engineering domains. Originally, ICA was designed to solve continuous optimization problems. This paper presents a binary version of ICA, dubbed ICA with Binary-encoding (ICAwB), to solve selection problems. ICAwB works with binary encoding and utilizes new socio-politically inspired operators. Additional features are incorporated within ICAwB to develop an improved version dubbed IICAwB. ICAwB and IICAwB with other binary versions of ICA are compared. IICAwB shows much better performance than existing binary ICAs and ICAwB. The proposed IICAwB is quite generic, and its applicability to other combinatorial optimization problems can be attempted with advantage.

Autism Spectrum Disorder is a neurological disorder linked to brain development that impacts facial features. An extensive and intricate neurodevelopmental disorder, ASD first appeared in early childhood. For healthcare professionals to treat and care for patients in a timely and appropriate manner, early recognition of ASD is essential. Many machine learning algorithms have been explored to investigate the viability of diagnosing autism. But finding accurate and timely ways to identify autism is still quite difficult. To improve autism identification accuracy while reducing time consumption, a new method termed Radial Adaptive Feature Projection based Generalized Emphasis Boost Classification (RAFP-GEBC) is presented. The primary goal of the RAFP-GEBC technique is to increase the accuracy of autism identification by means of effective processing. In order to identify autism spectrum disorder, this technique uses EEG signals from a dataset and includes pre-processing, feature selection, and classification. The Radial Basis Kernel Adaptive Stromberg Wavelet Filtering approach is used in the pre-processing stage. Input EEG signals are cleaned, transformed, and arranged into an appropriate manner using this technology. EEG signals are broken down into discrete frequency components, and noise is removed from each component in turn. Contingency Correlative Projection Pursuit Regression is then used in the feature identification process. The most pertinent and instructive characteristics are found through this procedure to ensure an appropriate classification of autism. The suggested RAFP-GEBC technique's feature selection cuts down on the amount of time needed for autism identification. The time needed to detect autism is decreased by the GEBC approach. In conclusion, the Generalized Learning Vector Quantized Emphasis Boost method is used to classify data with distorted features. By using an ensemble machine learning technique called “boosting,” classification results are strengthened and patients with and without autism can be distinguished with the least amount of error. As a result, the RAFP-GEBC method delivers precise and error-free autism identification. Numerous factors are experimentally evaluated by many people. According to qualitative study, the RAFP-GEBC strategy outperforms other approaches in the detection of autism.

The degradation of natural water resources, including rivers, reservoirs, and lakes, represents one of the most pressing environmental challenges today. Effective water quality management is essential to ensure sustainable utilization of these vital resources. Conventional machine learning methods often face limitations such as sparse and irregular sampling, as most water quality monitoring stations record data infrequently, typically on a monthly basis. Additionally, traditional optimization algorithms relying on random partitioning and cross-validation can produce imbalanced sample distributions, resulting in suboptimal prediction performance during testing. To address these challenges, this study proposes a novel Hybrid Whale Optimization with Long Short-Term Memory and Attention Mechanism (HWOA-LSTM-Attention) framework for accurate water quality forecasting. The framework leverages LSTM networks to capture temporal dependencies and incorporates an attention mechanism to assign adaptive weights to critical features, thereby enhancing predictive accuracy for complex and nonlinear water quality parameters. The Hybrid Whale Optimization Algorithm (HWOA) is employed to fine-tune model hyperparameters, optimizing performance metrics such as Mean Absolute Percentage Error (MAPE), Root Mean Square Error (RMSE), Absolute Proportion Error (APEmax), and the coefficient of determination (R²). Experimental results show that the proposed HWOA-LSTM-Attention framework achieves a high prediction accuracy of 96.84%, outperforming existing benchmark models. The approach enables water management authorities to forecast pollution levels more effectively, supporting early warning systems, disaster prevention, and real-time monitoring of pollutant dispersion across extensive water supply networks. This framework thus provides a robust, data-driven solution for sustainable and proactive water quality management.