http://repositorio.ufc.br/handle/riufc/478 2026-08-16T00:33:24Z http://repositorio.ufc.br/handle/riufc/87191 Título: Independent vector analysis with flexible statistical modeling for multimodal data fusion: from blind source separation to misinformation detection Autor(es): Damasceno, Lucas de Paula Abstract: The widespread dissemination of digital information through various communication channels has significantly altered the way societies generate, consume, and interpret knowledge. However, this shift has also accentuated the spread of misinformation. The intricate nature of this phenomenon lies in its inherently multimodal aspect, in which altered images, misleading texts, and synthetic media intertwine, forming false narratives that defy both human and machine comprehension. This thesis investigates the urgent issue of automated identification of multimodal misinformation based on a theoretical framework grounded in Independent Vector Analysis (IVA) and statistical signal analysis. First, we review the theoretical foundation of IVA, understood as an extension of Independent Component Analysis (ICA), which expands its ability to model multiple data sets by investigating the interdependencies between modalities. Based on this foundation, we propose a multivariate density estimation methodology based on the Maximum Entropy Principle (MEP), which combines global and local constraints through the Multivariate Entropy Maximization with Kernels (M-EMK) estimator. This estimator offers adaptive and expressive probability density functions, ensuring computational efficiency through Quasi-Monte Carlo integration, resulting in the development of the IVA-M-EMK algorithm. Besides source separation, this thesis applies the proposed approach to identifying multimodal misinformation, in which various data types, including text, images, and semantic embeddings, are analyzed together to highlight underlying consistencies and contradictions. Recognizing that misleading signals often exhibit sparse and organized characteristics, IVA-SPICE is employed, which integrates structured sparsity through inverse covariance estimation, enabling the efficient processing of high-dimensional multimodal spaces. Experimental investigations reveal that the proposed framework outperforms traditional unimodal and simplified fusion methodologies across multiple datasets, achieving greater accuracy, generalizability, and explainability. This thesis establishes Independent Vector Analysis, combined with adaptive density modeling and structured sparsity, as a solid, scalable, and interpretable foundation for multimodal data integration. Tipo: Tese 2025-01-01T00:00:00Z http://repositorio.ufc.br/handle/riufc/87102 Título: Discriminant Independent Vector Analysis Autor(es): Maia, Marília Magalhães Abstract: With the rapid advancement of technology and the accelerated growth in data production, Blind Source Separation (BSS) methods have gained increasing relevance due to their broad applicability across diverse domains. In scenarios where multiple data modalities are mixed and the objective is to recover the original underlying sources, techniques capable of exploiting relationships among them become essential, particularly in representation and classification tasks. The traditional method for handling such multimodal data is Independent Vector Analysis (IVA); however, its non-discriminative nature limits its performance when source separation is directly linked to classification objectives. In this context, this work introduces Discriminant Independent Vector Analysis (DIVA), a supervised extension of IVA constructed through the incorporation of Fisher’s Linear Discriminant (FLD) criterion into the IVA framework. The resulting method aims to estimate independent sources that, in addition to satisfying statistical independence, maximize class separability, making it particularly suitable for binary classification problems. The proposed model was implemented based on IVA-G, a widely established formulation in the literature. To evaluate the performance of DIVA-G, a Support Vector Machine (SVM) classifier was employed in a semi-supervised setting, using the F1-score as the primary evaluation metric. Experiments with synthetic datasets enabled the identification of statistical characteristics that favor the method, demonstrating consistent and superior performance compared with other IVA-derived algorithms. Subsequently, the real-world datasets NSL-KDD and MediaEval2016 were used to investigate the behavior of the method in complex and noisy scenarios. The results indicate satisfactory performance, stability, and reliability, suggesting that DIVA-G is a promising approach for discriminative source separation and warrants further, more in-depth investigation. Tipo: Dissertação 2026-01-01T00:00:00Z http://repositorio.ufc.br/handle/riufc/86964 Título: A multi-agent traceable semantic graph architecture for digital forensic knowledge representation and relational inference Autor(es): Monteiro, Marcos José Alves de Barros Abstract: Digital forensic investigations increasingly depend on the interpretation of heterogeneous evidence recovered from multiple computational artifacts. Although extraction tools have advanced considerably, connecting recovered traces to investigative reasoning remains a technical challenge, particularly when provenance and traceability must be preserved for later examination. This dissertation introduces a graph-centered multi-agent framework for organizing digital forensic evidence as a structured semantic representation linked to artifacts extracted from forensic disk images. The framework connects entities, semantic relations, provenance metadata, and investigative hypotheses within a unified Digital Forensic Knowledge Graph, enabling relational inference under incomplete-evidence conditions. Experimental evaluation showed 73.3% concept recovery, 73.3% semantic relation recovery, 57.0% hypothesis coverage, and full traceability coverage for evidentiary relations. Among the evaluated inference models, node2vec_b f s_ppmi_negative_l2 achieved the strongest overall performance under controlled perturbation settings. The findings show that graph-based semantic representation can support forensic reasoning while maintaining explicit linkage between inferred relations and traceable digital evidence. Tipo: Dissertação 2026-01-01T00:00:00Z http://repositorio.ufc.br/handle/riufc/86946 Título: Análise de falhas de coprocessadores compartilhados em um MPSoC para estender o ISA RISC-V Autor(es): Reis, Jorge Luiz Costa Abstract: Reduced Instruction Set (RISC) architectures optimize a complex ISA by implementing only the most frequently used instructions in hardware. However, the application execution time significantly increases when executing heavily used instructions in software. One technique that optimizes the trade-off of implementation cost and execution time is the use of a Multiprocessor System-on-Chip (MPSoC), in which processors extend their ISA by sharing coprocessors that implement lesser-used instructions. In this work the impact of shared coprocessor failures on two RISC-V MPSoC architectures is analyzed. In the first phase, we evaluated these architectures using two image processing applications and four different models of failure rates in terms of power dissipation, energy consumption, area consumption, maximum operating frequency, and execution time. The experiments in this phase show a maximum increase of 16% in execution time for the application with a lower percentage of instructions executed on the coprocessor. For the application with the highest rate of coprocessor use, the execution time does not increase significantly in the proposed architectural configurations for the MPSoC after the first failure scenario. For the second phase of the work, the MPSoC was expanded to allow testing with six different models and four failure rates. The experiments in this phase showed a 34% increase in execution time for the application with a lower rate of instructions executed on the coprocessor. For the other application, the increase reached 92% in execution time. Tipo: Dissertação 2023-01-01T00:00:00Z