Publications

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.

Magnetic Resonance Imaging (MRI) plays a critical role in the accurate and early detection of brain tumors, enabling effective clinical decision-making and treatment planning. However, distinguishing between malignant and benign tumors remain chal- lenging due to substantial variations in shape, texture and intensity patterns. This study proposes a hybrid deep learning frame- work that integrates Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM) networks to improve the accuracy of brain tumor classification from MRI scans. The model combines the spatial feature extraction of CNNs, the sequential pattern recognition of RNNs, and the long-range dependency modeling of LSTMs, creating a more robust and holistic classification system. The proposed approach was evaluated using the Kaggle Brain Tumor MRI dataset, comprising 3,000 labeled MRI images classified into tumor and non-tumor categories. The proposed model achieved an accuracy of 93%, precision of 91%, recall of 92%, F1-score of 91%, and an AUC of 0.95, outperforming single models such as CNN (91%). RMSprop was identified as the optimal optimizer, providing faster convergence and improved gradient sparsity. An ensemble meta-learning stacking strategy was employed to integrate CNN, RNN, and LSTM predictions, effectively reduc- ing misclassification and enhancing overall performance. These results highlight the potential of ensemble deep learning for improving diagnostic accuracy in clinical practice. Future work will focus on optimizing computational efficiency, incorporating additional imaging modalities, and enhancing interpretability to facilitate clinical adoption.

This paper presents a framework for synthesizing pathological speech by incorporating changes in the articulatory domain. Through a pilot study comparing GRU and LSTM architectures with an Adam optimizer, it was found that GRU outperformed LSTM in this context, though no general consensus exists on the superiority of one over the other for specific datasets. The framework was tested on MFCC (Mel- frequency cepstral co-efficient) predictions, where speaker-independent architectures demonstrated superior performance on the MOCHA-TIMIT datasets compared to speaker-dependent ones. An analysis of neural network activations revealed that these models learn line-like boundaries potentially representing phonetic structures. The study identified that the quality of synthesized speech is more limited by the vocoder than by the acoustic mapping, with 88% of the variance attributed to vocoder analysis-resynthesis. Furthermore, a statistically significant improvement in generalization performance was observed with the addition of more training data. The study also explored the synthesis of pathological speech, with informal feedback suggesting the synthesized examples resembled disordered speech. The framework offers a platform for implementing pathological articulations by mapping physiological changes in a low-dimensional articulatory space. Future work will focus on enhancing vocoder quality and developing models that consistently reflect specific speech pathologies. An open-source repository has been provided to reproduce these results.

Real-world problems often exhibit complex, nonlinear relationships between variables, which cannot be adequately addressed through linear assumptions. This paper explores how advanced algorithms, specifically the kernel trick, transform these nonlinear problems into higher-dimensional spaces, facilitating easier solutions. The mathematical underpinnings of the kernel trick, including the Volterra filter and Hilbert space transformations, are explored. The study implements Support Vector Machines (SVM) with Gaussian kernels to solve the XOR Toy problem, double Fibonacci spiral, and the Breast Cancer Wisconsin dataset, highlighting the efficacy of nonlinear transformations. Experimental results demonstrate high classification accuracy, validating the robust application of kernel methods in complex data analysis.

Nano-particles (NPs) are widely recognized as significant elements in a wide spectrum of goods, including aesthetics and electronics. Their application is expanding, despite the fact that their substantial financial and social possibilities has yet to be fulfilled. NPs possess distinct features that make them helpful in a range of purposes; nonetheless, their apparent toxicity increases security problems. The novelty of the study is autonomous detection of nanoparticles with improved surveillance. Advances have been done to comprehend the threats that NPs represent to human well-being and the surroundings, but further study and surveillance are required. In the past decade, Machine Learning (ML) approaches have used massive databases and computational capacity to make advancements in domains ranging from recognizing faces to genetics. In recent years, ML approaches are being used in nanotoxicology, with very promising findings. This study used ML techniques to autonomously detect NPs depending on their physical features, resulting in excellent classification accuracy. The findings show that ML algorithms are efficient in detecting nanoparticles and emphasize the requirement for more accurate characterization approaches to assure their safety in a variety of uses. The proposed ML algorithms was found to be efficient at detecting NPs, with the NN method outperforming all others with an accuracy of 0.95.

Existing techniques often depend on manual processes or limited models for machine learning that do not capture complex, multimodal dependencies between code, infrastructure and runtime behaviour. This paper presents an AI-driven approach using Large Language Models (LLMs) for efficient bug triaging and proactive root cause analysis. Our system integrates graph-based modelling, multimodal learning, and LLMs to automatically triage bugs, predict root causes, and generate effective test cases for bug reproduction. We evaluate the proposed framework’s effectiveness using various basic models from existing literature, including traditional machine-learning classifiers and graph-based models. The results show that our method is significantly higher than the baseline approaches in the average time to resolution (MTTR), the accuracy of the root cause (RCA) and the effectiveness of the test case (THE), achieving a reduction of MTTR by more than 50%, an RCA of 84.7% and a THE of 83.2%. Our approach improves the efficiency and accuracy of bug triaging and software quality assurance by generating reliable test cases.

We introduce Code Prophet, a new framework powered by large language models (lLMs) that transforms highlevel problem statements into structured, sprint-ready software development plans. Unlike traditional tools dependent on explicit input, Code Prophet provides indicated and latent requirements by combining GPT-3.5, a conditional Generative Adversarial Network (GAN), and contextual trend analysis. The multi-stage architecture includes a GPT-3.5 + GAN fusion for user story generation, a Graph Neural Network (GNN) for dependencyconscious task planning, and a Boost-based time estimate with Sentence-BERT-driven developer profiling. Trained on a rich, multimodal annotated dataset that supports Code Prophet’s realtime, interactive planning via IDE-integrated agents. Experimental results show significant gains in story completeness, contextual relevance, and planning precision over baselines, positioning Code Prophet as a forward-looking co-creator in the software engineering lifecycle.

Predicting water quality using machine learning (ML) and artificial intelligence (AI) techniques has emerged as a key tactic for preserving public health and managing water resources sustainably. Conventional techniques for QW assessment are usually labour-intensive, time-consuming, and reactive. On the other hand, by utilising vast datasets from sensors, satellite imaging, and lab results, ML and AI approaches allow for automated, real-time, and predictive study of water quality. These models anticipate important water characteristics like turbidity, pollutant levels dissolved oxygen, and pH using supervised, unsupervised, and hybrid learning techniques. In the forecast of quality of water, methods such as deep learning models have demonstrated encouraging accuracy. AI-powered models are also able to pinpoint the causes of pollution and suggest prompt fixes. This study provides an assertive, flexible, and affordable approach to water resource management by highlighting the importance, methods, and uses of ML and AI in forecast of QW.