Intel (Brazil)

The rapid growth of IoT devices demands scalable and efficient threat detection solutions. This paper introduces a sample-based, multi-stage machine learning (ML) pipeline for IoT threat detection using the CICIoT2023 dataset, integrating feature selection, data balancing, and hyperparameter optimization to improve detection accuracy while reducing the computational overhead associated with training. We evaluate, and across binary, multiclass, and fine-grained tasks, showing that with 10% sampling achieves the best trade-off between accuracy and efficiency. Compared to prior methods, our approach eliminates GPU dependence, maintains low latency, and preserves state-of-the-art performance while enabling scalable training for high generalization capacity. Additionally, we provide model selection guidelines based on dataset complexity and computational constraints. The results show that training with a sample-based approach enables effective threat detection on large datasets, producing models that generalize well to diverse IoT attack scenarios, thus ensuring practical applicability in real-world deployments.

This paper focuses on the scenario of industrial communication protocols, with observance for the transition from fieldbus networks to protocols subsidized to the Ethernet standard. The study emphasizes the Ethernet – Advanced Physical Layer (Ethernet-APL), which enables the use of Ethernet protocols in diverse industrial environments, providing high bandwidth capacity and longer cable lengths, along with the ability to transmit both power and data through the same cabling. Thus, the work aims to develop a practical application in which there is a Process Field Network (PROFINET) network that enables the connection of a temperature sensor connected by Ethernet-APL, communicating with industrial PROFINET elements. Temporal network communication metrics will also be analyzed.

The production of electronic boards is a common activity in the industrial environment, and ensuring their quality is essential for obtaining reliable products. This work presents a comparative study of the performance of three machine learning architectures: YoloV8, FOMO, and MobileNet. The dataset was created based on four classes for assembly fault detection. The model that achieved the best results was FOMO, with precision, recall, and Fl-Score above 95 %, as well as processing 10.5 frames per second and having a total size of 152 kB.

This work presents a reconfigurable metasurface for frequency and polarization control at S-band microwave frequencies. The structure is based on unit cells composed of complementary split-ring resonators (CSRR), namely, metallic patches and cross-shaped grids with gaps. PIN diodes are aligned along the x-axis to connect the grid, enabling bandpass frequency shifting of x-polarized waves with 1-bit keying. Numerical simulations demonstrate high extinction ratios of up to 21.71 dB and cross-talk peaks of 20.28 dB, making this design promising for future wireless communications with 1-bit reconfigurable metasurfaces.

Countless issues of the daily life of municipalities have constantly concerned cities, which have been addressed in the last decades under the umbrella of Smart Cities (SCs) initiatives. However, the literature has pointed out that proposing and executing projects in SCs initiatives must be aligned to the Strategic Planning (SP) of the city. Indeed, tackling the different interests in each category of city issues and ensuring effective community participation for indicating city goals can be the greatest challenge for local public administration, as it requires technical, modern, practical, and effective SP. Brazil is a large country with populous and dense cities, which also present these issues but also have appeared in diverse SCs rankings. The ones that most appear in such rankings are usually the capitals of Brazilian states. However, is not clear whether all these cities have their SP and how citizens follow their evolution. So this work aims to investigate if Brazilian state capitals have well-defined SP and if they actually use it as a management tool for proposing smart projects for developing smart(er) cities. More specifically, 27 state capitals were analyzed and the results showed that only 48% of the Brazilian state capitals have formally constituted planning artifacts, mostly by normative impositions, but only 30% of them monitor their SP through indicators. So, in order to turn the cities smarter, more than investing in technology, it is also necessary to innovate in planning, tracking and monitoring their projects and indicators.

Objective: The objective of this article is to present a conceptual model that includes the elements and practices of Digital Transformation project management to support Business Agility, as the adoption of project management practices and digital technologies allows organizations to adapt flexibly to the business environment in which they operate. Method/design/approach: To achieve the objective, field research was carried out using interviews and focus groups. Results and conclusions: Based on the results, it was possible to present a conceptual model that represents the elements and practices of managing Digital Transformation projects to support Business Agility in organizations. Research Contributions: In relation to theoretical contributions, the results obtained from the literature review compared with empirical research allowed discussions to advance on the phenomenon of Digital Transformation and Business Agility. Originality/Value: A practical model capable of supporting project management professionals during the life cycle of Digital Transformation projects stands out.

No Brasil, a ausência de mecanismo institucional que normatize a interoperabilidade informacional entre o Sistema Brasileiro de Inteligência (Sisbin) e o Ministério Público, bem como a aplicação pelo órgão ministerial dos produtos oriundos da Atividade de Inteligência impede a plena utilização do saber estatal na promoção do interesse nacional e restringe o combate às principais ameaças que colocam em risco a integridade da sociedade e do Estado. O presente artigo, com base em pesquisa e análise documental-legislativa e jurisprudencial,coletou e sistematizou diplomas legais e julgados concernentes ao intercâmbio de dados e Conhecimentos entre determinadas agências integrantes do Sisbin e o Parquet, a fim de caracterizar o estágio atual desta relação.O levantamento apontou que o arcabouço normativo-legal vigente carreia prescrições que, usualmente: (I) não abarcam as relações envolvendo o Parquet e as 48 agências do Sisbin em sua totalidade e; (II) são genéricas e esbarram em questionamentos de natureza procedimental acerca do emprego dos dados e Conhecimentos, bem como da atuação cotidiana de profissionais de Inteligência e membros do órgão ministerial. A regulação da interoperabilidade informacional Sisbin-Ministério Público dotaria estas instituições de instrumentos para atuação harmônica na promoção da segurança do Estado e da Sociedade, além de efetivar objetivos e diretrizes estratégicos consignados na Política Nacional de Inteligência e na Estratégia Nacional de Inteligência. Na França, o debate acerca da interoperabilidade informacional Atividade de Inteligência-Ministério Público progride, redundando na inserção de dispositivos em diplomas legais que balizam esta cooperação, tanto em seus aspectos procedimentais, temáticos e de fluxos.

Serviços de Inteligência produzem conhecimentos sobre conflitos desde que foram criados e institucionalizados. A natureza dessas disputas, entretanto, sofreu transformações ao longo do tempo, principalmente após as guerras mundiais do século XX e, mais recentemente, com o fim da Guerra Fria. A análise de conflitos se desenvolveu nesse contexto, reunindo estudos sobre guerra e paz, diplomacia, negociação, prevenção e gerenciamento de conflitos. Este artigo objetiva explorar possibilidades de aproveitamento de modelos, técnicas e ferramentas da análise de conflitos em apoio ao processo de produção de conhecimentos de Inteligência, com vistas a contribuir para a discussão, ainda incipiente, sobre convergências e divergências entre as duas práticas. Para isso, foram analisados e cotejados estudos e manuais práticos sobre atividade de Inteligência, análise de conflitos e produção de conhecimento. Para além da apresentação de ferramentas e possíveis usos práticos de técnicas acessórias, o artigo também incentiva a reflexão dos profissionais de Inteligência a respeito de seus objetos de análise enquanto espaços de disputa e competição entre diferentes partes.

The financial fraud risk assessment requires expertise concerning the audit methodology and the risk assessment of business processes. The assessment of financial fraud risks is a significant challenge for independent, external auditors chiefly when dealing with the audit in an information technology (IT) environment. A meta-analysis financial fraud risk governance model taking into account an IT environment and based on fuzzy inference system is proposed in this paper. The fuzzy set theory and fuzzy logic are employed to deal with actual business audits, which do not always concern dichotomic fraud risk conditions. The inputs of the proposed fuzzy financial fraud risk governance system concern meta-analysis input factors related to the fundamental domains of risk in IT audit comprising the (i) effective identity and access protocol, (ii) system development, (iii) control of business operations, and (iv) change in systems or applications. Results demonstrate that the proposed approach enables supporting the operational risk management to promote the operational efficiency by identifying, measuring, and disclosing events (risk conditions) both in terms of qualitative (stratification) and quantitative (score) analysis. The fuzzy IT financial fraud risk system is able to work as a first barrier to reflect the adequacy of the information technologies and systems used to avoid risk in the financial fraud governance.

The COVID-19 pandemic generated research interest in automated models to perform classification and segmentation from medical imaging of COVID-19 patients, However, applications in real-world scenarios are still needed. We describe the development and deployment of COVID-19 decision support and segmentation system. A partnership with a Brazilian radiologist consortium, gave us access to 1000s of labeled computed tomography (CT) and X-ray images from São Paulo Hospitals. The system used EfficientNet and EfficientDet networks, state-of-the-art convolutional neural networks for natural images classification and segmentation, in a real-time scalable scenario in communication with a Picture Archiving and Communication System (PACS). Additionally, the system could reject non-related images, using header analysis and classifiers. We achieved CT and X-ray classification accuracies of 0.94 and 0.98, respectively, and Dice coefficient for lung and covid findings segmentations of 0.98 and 0.73, respectively. The median response time was 7 s for X-ray and 4 min for CT.