Complex graph neural networks for medication interaction verification

Westarb, Gustavo; Stefenon, Stéfano Frizzo; Hoppe, Aurélio Faustino; Sartori, Andreza; Klaar, Anne Carolina Rodrigues; Leithardt, Valderi Reis Quietinho

doi:10.3233/jifs-223656

Cited by 7 publications

(3 citation statements)

References 45 publications

(28 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Leveraging artificial intelligence (AI) techniques emerges as an enticing approach to addressing this issue. AI, a domain within computer science, demonstrates the capacity to comprehend tasks [134]; analyze data [135]; evaluate time series [136], optimization design [137], and classification tasks [138][139][140]; and make decisions using algorithms crafted by experts [141].…”

Section: Artificial Intelligence Applicationsmentioning

confidence: 99%

A Review of Automation and Sensors: Parameter Control of Thermal Treatments for Electrical Power Generation

Buratto,

Muniz,

Nied

et al. 2024

Sensors

View full text Add to dashboard Cite

This review delves into the critical role of automation and sensor technologies in optimizing parameters for thermal treatments within electrical power generation. The demand for efficient and sustainable power generation has led to a significant reliance on thermal treatments in power plants. However, ensuring precise control over these treatments remains challenging, necessitating the integration of advanced automation and sensor systems. This paper evaluates the pivotal aspects of automation, emphasizing its capacity to streamline operations, enhance safety, and optimize energy efficiency in thermal treatment processes. Additionally, it highlights the indispensable role of sensors in monitoring and regulating crucial parameters, such as temperature, pressure, and flow rates. These sensors enable real-time data acquisition, facilitating immediate adjustments to maintain optimal operating conditions and prevent system failures. It explores the recent technological advancements, including machine learning algorithms and IoT integration, which have revolutionized automation and sensor capabilities in thermal treatment control. Incorporating these innovations has significantly improved the precision and adaptability of control systems, resulting in heightened performance and reduced environmental impact. This review underscores the imperative nature of automation and sensor technologies in thermal treatments for electrical power generation, emphasizing their pivotal role in enhancing operational efficiency, ensuring reliability, and advancing sustainability in power generation processes.

show abstract

Section: Artificial Intelligence Applicationsmentioning

confidence: 99%

A Review of Automation and Sensors: Parameter Control of Thermal Treatments for Electrical Power Generation

Buratto,

Muniz,

Nied

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…In addition, the cultural, institutional, and technical aspects can also be added, depending on the complexity and need foreseen, as well as the size of each project [66][67][68]. Nowadays, AI-based models are applied considering their capacity to deal with nonlinear data [69][70][71].…”

Section: Technology and Efficiency Improvementmentioning

confidence: 99%

The Sustainability Concept: A Review Focusing on Energy

Muniz,

da Costa Júnior,

Buratto

et al. 2023

Sustainability

View full text Add to dashboard Cite

The concept of sustainability, with a focus on energy, has emerged as a central tenet in addressing the mounting global challenges of environmental degradation and resource depletion. Indicators of sustainability focusing on energy are crucial tools used to assess and monitor progress toward achieving a more sustainable energy system. These indicators provide valuable insights into the environmental, social, and economic dimensions of energy practices and their long-term impacts. By analyzing and understanding these indicators, policymakers, businesses, and communities can make informed decisions, formulate effective policies, and steer their efforts toward a more sustainable energy future. These indicators serve as navigational guides, steering the world toward energy practices that support both present needs and the well-being of future generations. In this paper, the concept of sustainability and measurement indexes used are reviewed, focusing on energy factors. The focus of the discussion presented here is related to an assessment of the possibilities for improving energy efficiency and evaluating the indicators that are used to measure whether the desired levels of sustainability are being achieved.

show abstract

“…Graphs, defined as a set of nodes (vertices) and a set of links (edges) that connect some of those vertices, constitute a very useful tool to represent networks. The application of graphs has received thorough attention from the scientific community since its application is open to a wide number of research areas, with graph neural networks being a recent hot topic [24,25]. Additionally, graph coloring represents another hot topic in the field of discrete mathematics, attracting the interest of the scientific community from the fields of both engineering and mathematics, due to its theoretical challenges and applications [26].…”

Section: Graphs and Channel Selection In Wi-fimentioning

confidence: 99%

Channel Selection in Uncoordinated IEEE 802.11 Networks Using Graph Coloring

Gimenez-Guzman,

Marsa-Maestre,

de la Hoz

et al. 2023

Sensors

View full text Add to dashboard Cite

One of the big challenges in decentralized Wi-Fi networks is how to select channels for the different access points (APs) and their associated stations (STAs) in order to minimize interference and hence maximize throughput. Interestingly enough, de facto standards in terms of uncoordinated channel selection are quite simple, and in many cases result in fairly suboptimal channel allocations. Here, we explore how graph coloring can be used to evaluate and inform decisions on Wi-Fi channel selection in uncoordinated settings. Graph coloring, in its most basic form, is a classic mathematical problem where colors have to be assigned to nodes in a graph while avoiding assigning the same color to adjacent nodes. In this paper, we modeled Wi-Fi uncoordinated channel selection as a graph coloring problem and evaluated the performance of different uncoordinated channel selection techniques in a set of representative scenarios of residential buildings. The results confirm some of the widely accepted consensus regarding uncoordinated channel selection but also provide some new insights. For instance, in some settings, it would be better to delegate the decision on which channel to use to transmit the STAs, rather than having the AP make the decision on its own, which is the usual way.

show abstract

Complex graph neural networks for medication interaction verification

Cited by 7 publications

References 45 publications

A Review of Automation and Sensors: Parameter Control of Thermal Treatments for Electrical Power Generation

A Review of Automation and Sensors: Parameter Control of Thermal Treatments for Electrical Power Generation

The Sustainability Concept: A Review Focusing on Energy

Channel Selection in Uncoordinated IEEE 802.11 Networks Using Graph Coloring

Contact Info

Product

Resources

About