As the portfolio optimization field grows, classical techniques often notoriously find it difficult to efficiently model how investors decisions, risk tolerances, and asset attributes intertwine. This paper presents an innovation-based hybrid method, where Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) combined with Additive Ratio Assessment (ARAS) for multi-criteria decision making, Goal Programming (GP) and a Genetic Algorithm (GA) for finding constraints are united. The proposed approach enhances the accuracy of ranking and effectiveness of allocation by incorporating asset evaluation, characterization of investors and probabilistic construction of portfolios. The system is tested in view of various performance implications, using the FAR-Trans dataset, a collection of genuine transaction statistics and asset pricing, as well as investor data. The first step involves project transaction capacities partitioning and risk categorization to create a bipartite TOPSIS–ARAS scoring mechanism. The GP part of the model matches investment decisions to the individual return and risk expectations of each investor, and the GA promotes the use of entropy-aware strategies. Important performance metrics are a Sharpe Ratio of 2.241, the annualized return of 4.6% and diversification score of 0.845. The study also reflects a 0.729 correlation between TOPSIS–ARAS rankings, and GP configurations leading to portfolio returns of over 30.0%. The system offers a realistic depiction of the behavior of investors, considering several transaction channels and different risk factors as well as geographies. The comprehensive integration is very flexible, computationally effective and based on realistic investment models while minimizing constraint deviation.

Diabetes, afflicting 537 million worldwide, is a prevalent and lethal non-communicable ailment. Its onset, influenced by factors like obesity and family history, manifests symptoms such as frequent urination. Long-term complications encompass heart, kidney, and nerve ailments. Early prediction mitigates risks. All these encompassing strategies are designed to improve prediction precision and facilitate proactive diabetes control. This research employed SMOTE methods to tackle imbalanced classes, utilizing various classification algorithms such as Random Forest, XGBoost, Multilayer Perceptron, Gradient Boost, and AdaBoost. Following extensive training and evaluation, the AdaBoost classifier delivered superior outcomes, achieving a 94.02% accuracy rate, an F1 score of 93.32%, and an AUC of 0.95. In the healthcare industry, accurately forecasting diabetes mellitus is crucial; however, privacy laws hinder the transfer of medical information from the Internet of Medical Things (IoMT), causing delays in diagnosis. This study introduces the Federated Learning with Weighted Conglomeration Optimization (FLWCO) model as a solution to these challenges. In Centralized Learning, AdaBoost with WCO achieves an accuracy of 95.32% when tested on a Kaggle dataset consisting of 96,146 instances. During the second stage, FLWCO achieves a superior 97.27% accuracy rate compared to other federated learning techniques. The method not only guarantees privacy conformity but also decreases communication expenses. FLWCO demonstrates superiority over existing federated learning algorithms in real-world heart illness prediction. Furthermore, the proposed model can be employed to estimate the likelihood of heart disease in individuals with diabetes. This highlights the potential of federated learning, especially FLWCO, in leveraging distributed data while preserving privacy, facilitating accurate diabetes mellitus diagnosis, and addressing challenges in sharing medical information securely and efficiently.

Diabetic foot ulcer (DFU) is one of the most expensive and debilitating complications of diabetes. According to procedures developed by the National Institute of Health and Clinical Excellence, early and effective treatment of DFU can minimize the severity of complications, such as unnecessary amputations, and improve overall quality of life. In this paper, a stacked parallel convolution layers based Convolution Neural Network (CNN) is proposed, known as DFU MobileNet, to classify skin infected with DFU against normal skin. The main objective of this work is to develop a DFU recognition system that accurately identifies infected skin. DFU MobileNet consists of three blocks of parallel convolution layers, each with a unique kernel size to extract local and global features distinctly. Using the DFUC 2020 dataset, the proposed DFU MobileNet outperformed existing state-of-the-art models, achieving an accuracy of 92.2% using CNN DenseNet, 95.4% using MobileNet, and 97.8% when combining DenseNet and MobileNet.

Deep learning and self-supervised learning techniques have advanced, making it possible to diagnose medical images more accurately. Our goal in this work is to increase the accuracy of medical diagnosis using chest X-rays by utilising information distillation and model compression techniques. By extracting knowledge from a self-supervised model which is a teacher model (SWAV Model with ResNet-50 as backbone), we enhance inference speed and reduce the computational resources needed for correct diagnosis. Our strategy entails translating the knowledge acquired by self-supervised models to smaller, more efficient models without sacrificing accuracy. In addition, we look at how model compression and distillation affect the diagnosis’s interpretation. The results of this study may enhance medical diagnosis procedures and increase their accessibility in environments with limited resources. Our extensive experiments prove the efficacy with 97.34% accuracy for the student model with knowledge distillation.

The exponential growth of user-generated content on social media platforms has increased the diffusion of hate speech and fake information considerably, which poses serious risks to the integrity of societies and the public trust people vest in them. Existing approaches tend to be critically limited by core restrictions such as unimodal processing, absence of user-context integration, poor cross-modal synchronizations, and weak robustness against adversarial manipulations. Such restrictions lead compounded losses in generalizations, misclassifications, and unreliability of deployments in real-life scenarios. For these challenges, this work offers a complete, modular, and analytically validated pipeline around five newly developed methods for multimodal hate speech and fake post detection from a Twitter corpus. The first component is Context-Driven Social Vector Extraction, which produces enriched contextual embeddings by capturing text, image, temporal, and behavioural metadata. These embeddings are processed by the Cross-Modal Label Fusion via Mutual Co-Attention (CMF-MCA) module, which utilizes a dual-transformer with co-attention to jointly annotate textual and visual information. The third module brought into play is Semantic Propagation Graph for Hate & Fake Correlation (SPG-HFC), which implements a multi-relational graph attention mechanism that captures semantic influence and community-level propagation patterns. The next step is located in the Adaptive Modality Routing via Reinforcement (AMR-R) module which selects, depending on the input complexity, between machine learning and deep learning classifiers using a reinforcement policy network. Finally, the Counterfactual Consistency Validation Engine (CCVE) performs post-prediction verification by evaluating counterfactual variants to ensure robustness and semantic consistency. Thus, detection accuracy, scalability, and inference reliability significantly increase in this integrated framework. It transcends the realm by making permissible cross-modal, context-aware, and behaviour-driven classification, which in turn greatly supports the development of large-scale secure and trustworthy content moderation systems in process.