Navegando por Autor "Andrade, Ermeson Carneiro de"

Agora exibindo 1 - 15 de 15

A comprehensive software aging analysis in LLMs-based systems
(2025) Santos, César Henrique Araújo dos; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/9618931332191622
Large language models (LLMs) are increasingly popular in academia and industry due to their wide applicability across various domains. With their rising use in daily tasks, ensuring their reliability is crucial for both specific tasks and broader societal impact. Failures in LLMs can lead to serious consequences such as interruptions in services, disruptions in workflow, and delays in task completion. Despite significant efforts to understand LLMs from different perspectives, there has been a lack of focus on their continuous execution over long periods to identify signs of software aging. In this study, we experimentally investigate software aging in LLM-based systems using Pythia, OPT, and GPT Neo as the LLM models. Through statistical analysis of measurement data, we identify suspicious trends of software aging associated with memory usage under various workloads. These trends are further confirmed by the Mann-Kendall test. Additionally, our process analysis reveals potential suspicious processes that may contribute to memory degradation.
A modeling-based approach for dependability analysis of a constellation of satellites
(2024-02-28) Farias, Daniel Castro de; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/5992660907313176
Satellite constellations play critical roles across various sectors, encompassing communication, Earth observation, and space exploration. Ensuring the dependable operation of these constellations is of utmost importance. This paper introduces a dependability modeling approach using stochastic Petri nets to analyze satellite constellations. The primary focus is on improving operational efficiency through the assessment of availability, reliability, and maintainability. The approach helps satellite designers make informed decisions when selecting constellation configurations by assessing various dependability metrics. Using a global navigation satellite system as a case study, we conduct extensive numerical experiments to evaluate the feasibility of our approach. The results demonstrate quantitatively the significant impact of redundant components on both reliability and availability. They also illustrate how utilizing satellites in repair and operational orbits can influence these metrics and highlight the direct correlation between reliability and maintainability.
Analisando o Backup-as-a-Service como uma estratégia de recuperação de desastres
(2021-06-02) Queiroz, Ewerton Cleyton Silva de; Andrade, Ermeson Carneiro de; Mendonça Neto, Júlio Rodrigues de; http://lattes.cnpq.br/7849727159222731; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/1234353605805269
In modern environments, failures in information and communication technology (ICT) systems can have several consequences for a business, like data and revenue loss and customers dissatisfaction. Disaster recovery (DR) solutions, as BackupasaService (BaaS), has been adopted by companies as a way to avoid these problems and assure business continuity. Nevertheless, there are plenty of variables to consider during the adoption of a DR solution. Then, in this work, we present an integrated approach using experiments and models to evaluate a BaaS environment designed for DR. In our analysis, we consider relevant DR metrics like availability, downtime, RTO (Recovery Time Objective), and RPO (Recovery Point Objective). The results demonstrate that once BaaS is applied, the environment availability can vary according to the amount of data needed to be backed up or restored. Furthermore, sensitivity analysis indicates that the time needed to recover the data center and the backup interval are the most important parameter values for metrics like RTO and RPO. The proposed approach can help companies or individuals involved in the decisionmaking process for purchasing a DR solution.
Análise de envelhecimento de software em uma plataforma de Blockchain
(2022-05-04) Silva, Douglas Dias da; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/5082801636483279
Software aging is a phenomenon that plagues many long-running complex computer systems, which exhibit performance degradation or an increasing failure rate. Such a phenomenon may also be present in blockchain platforms. However, there are still no works focused on analyzing this phenomenon on these platforms. Thus, we adopted the Cardano blockchain to analyze software aging due to the presence of this technology in critical projects, its open-source nature and for being a sustainable solution. Considering the analysis of running a Cardano node on two computers with different configurations, we found evidence of software aging through memory degradation that was confirmed by the Mann-Kendall test. By analyzing the running processes, we confirmed that cardanonode (the main process of the platform) is the process possibly responsible for such degradation.
Análise de sentimentos de tweets relacionados a vacinas antes e durante a pandemia da COVID-19 no Brasil
(2023-03-01) Silva, Íkaro Alef de Lima; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/7938306473921402
In early 2020, the COVID-19 disease spread rapidly around the world and one of the ways to fight it is the vaccine. Governments faced problems with fake news and anti-vaccination groups. Thus, it is necessary to understand the feelings of the population in order to propose efficient public policies. This article describes a sentiment analysis on vaccine-related tweets in Brazil from June 2020 to June 2021. The results revealed peaks in total tweets in January and May 2021, the predominance of positive tweets, and feelings of confidence, fear, submission and sadness. They are also associated with former President Jair Bolsonaro. The negative polarity was the least common, showing that the Brazilian population was receptive to vaccines.
Análise de sentimentos dos tweets relacionados ao Superior Tribunal Federal no ano de 2019
(2022-11-10) Cadengue, Guilherme Lapa de Araújo; Andrade, Ermeson Carneiro de; Bocanegra, Silvana; http://lattes.cnpq.br/4596111202208863; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/8502533221842320
The Social media since its inception has affected all Internet users. Networks such as Twitter provide a new form of communication, interaction and, above all, a way of expressing opinions about the different events of life in society, consequently enabling the generation of content. Knowing the opinions of Brazilians about public institutions is very important for engaging people in society, as agents participating in decisions that affect all individuals, that is, it is a form of social inclusion. The application of Sentiment Analysis is carried out in several areas in order to extract the content of public opinion. The objective of this work is to identify the feelings of the Brazilian population about the Superior Federal Court of Brazil through the content of published tweets between January and December 2019. For this, the tweets in the period were collected, which were pre-processed, classified and then analyzed. The results show highly polarized opinions, but generally negative opinions regarding the STF are predominant (estimate at 51.7%).
Análise de sentimentos em Tweets relacionados ao desmatamento da Floresta Amazônica
(2021-12-17) Silva, Vinicius José Paes e; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/7437953784606274
The Amazon Forest is being devastated at the fastest pace in recent years. In 2021, the Amazon rainforest registers the largest accumulation of deforestation in 5 years, increasing from 13 thousand km2 between August 2020 and July 2021. An increase of 22% compared to the same period in the previous year, the highest number since 2006. Although many works address the issue of deforestation, none of them focus on analyzing the sentiments of the Brazilian population regarding the issue. This work presents an analysis of the sentiments of the Brazilian population related to the deforestation of the Amazon rainforest through the text mining of Twitter and aims to understand how Brazilian users opine and dialogue about the deforestation of the Amazon rainforest. The results reveal that Brazilian users tend to react to events related to deforestation in the Amazon forest on Twiter and that most users have a negative sentiment about the topic, reaching peaks of approximately 60% of tweets in a given time.
Análise de Sentimos de Tweets Relacionados ao Uso de Máscara Durante a Pandemia da Covid-19 no Brasil
(2022-10-07) Oliveira, Felipe de Araújo Morais Vilar; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972
The world has recently gone through a global crisis. The COVID-19 pandemic began in a Chinese city called Wuhan in mid-December 2019 and spread across the world, infecting more than 596 million people and causing about 6.68 million fatalities. As the COVID-19 virus has much of its proliferation and contagion through the airways, experts and scholars in the health area recommended that the entire population wear masks in an attempt to stop the number of cases by creating a physical barrier to try to contain the respiratory droplets that serve as a means of spreading the virus. The use of masks in Brazil was adopted at the beginning of April of the year 2020, but its mandatory only started around the end of May of the same year. However, the misinformation about the use of the face mask generated great controversy, doubts and discomfort among the Brazilian population. This work aims to analyze the feeling of the Brazilian population regarding the use of masks as PPE (Personal Protective Equipment) through posts (tweets) taken from Twitter. The results reveal that an average of 89.3% of the tweets related to face masks were neutral. Most of these neutral tweets show the Brazilian population’s discomfort in using masks, but at the same time accepting the need to use them in an attempt to stop the spread of COVID-19.
Aprendizagem colaborativa apoiada por computador durante a pandemia do Covid-19: uma revisão sistemática
(2022-10-07) Melo, Daniel Lemos de; Araújo, Carlos Julian Menezes; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/3156174527107999; http://lattes.cnpq.br/4495434877384543
With the social isolation measures caused by the pandemic in 2020, educational institutions had to plan alternatives to ensure remote teaching. Among the various approaches, we cite the Computer Supported Collaborative Learning (CSCL) strategy, which seeks, through computers, to promote collaborative teaching remotely. This study seeks to evaluate how the use of the CSCL was during this period of isolation, through a systematic review. From the query in several databases, it was possible to identify 120 works that reported experiences using CSCL. The results indicate considerable success in adopting the CSCL strategy, as well as trends in technology use and approaches in each region of the planet.
Avaliação experimental de replicação em banco de dados para recuperação de desastres
(2020-12-18) Santos Neto, Wilson Medeiros dos; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/1208163441814364
IT systems are essential for the operations of any modern business. Such systems must support operations of their corresponding company under any conditions. Disaster Recovery (DR) strategies have been implemented to help organizations mitigate unexpected failures and reduce unnecessary expenses. However, to the best of our knowledge, no other work experimentally analyzes data replication at the database layer with a focus on DR strategies. Therefore, this work evaluates a relational database replication as a mean of implementing a DR solution. We use a real testbed in a public cloud environment to perform extensive experiments aimed at implementing the replication provided by MySQL, considering various scenarios in the context of DR. Our results show how response time, Recovery Point Objective (RPO) and Recovery Time Objective (RTO) vary according to the size of the replicated data, the synchronization type (ex.: asynchronous or semisynchronous) and the configuration of the slave servers. This work can assist DR coordinators or individuals to decide which database replication configuration for disaster recovery is best for their work environment.
BRIMO: uma ferramenta para análise de sentimentos
(2022-04-08) Sales, Otávio Alves Guedes; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/9546787079254144
Twitter is one of the fastest growing social networks in recent years, being one of the main platforms for people to share their opinions through short posts called tweets. Every trending topic generates rich discussions, and an analysis of people’s sentiments about a topic provides a more comprehensive understanding of the subject. However, to our knowledge, there are no tools that make this type of analysis accessible to the general public. Thus, we propose BRIMO, a free web tool with simple and intuitive interface, which allows the analysis of sentiments quickly and objectively, through graphical representations of data. We evaluated BRIMO’s usability and usefulness with target users. The results indicate that the tool has great usability according to the criteria used, in addition to being useful for several purposes, in the perception of participants.
Desenvolvimento de jogos: Uma Abordagem Prática com Unity e Scrum
(2025-03-21) Silva, Cleydson Paes da; Araújo, Julian; Ricardo, Danilo; Andrade, Ermeson Carneiro de; http://lattes.cnpq.br/2466077615273972
Este trabalho apresenta um relato de experiência sobre o desenvolvimento de um jogo de survival horror utilizando a engine Unity, explorando os desafios técnicos e práticos encontrados no processo. O projeto combina conhecimentos téoricos e práticos da Ciência da Computação em diversas áreas, desde programação e design até gerenciamento de projetos. A metodologia Scrum foi adotada para organizar as tarefas, e ferramentas como Blender, GIMP, Audacity e Jira foram utilizadas em conjunto com a Unity. O artigo discute as principais dificuldades enfrentadas, as soluções implementadas e os aprendizados adquiridos, oferecendo uma referência valiosa para estudantes e futuros desenvolvedores de jogos.
Explainable Artificial Intelligence - uma análise dos trade-offs entre desempenho e explicabilidade
(2023-08-18) Assis, André Carlos Santos de; Andrade, Ermeson Carneiro de; Silva, Douglas Véras e; http://lattes.cnpq.br/2969243668455081; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/3963132175829207
Explainability is essential for users to efficiently understand, trust, and manage computer systems that use artificial intelligence. Thus, as well as assertiveness, understanding how the decision-making process of the models occurred is fundamental. While there are studies that focus on the explainability of artificial intelligence algorithms, it is important to highlight that, as far as we know, none of them have comprehensively analyzed the trade-offs between performance and explainability. In this sense, this research aims to fill this gap by investigating both transparent algorithms, such as Decision Tree and Logistic Regression, and opaque algorithms, such as Random Forest and Support Vector Machine, in order to evaluate the trade-offs between performance and explainability. The results reveal that opaque algorithms have a low explanability and do not perform well regarding response time due to their complexity, but are more assertive. On the other hand, transparent algorithms have a more effective explainability and better performance regarding response time, but in our experiments, we observed that accuracy obtained was lower than the accuracy of opaque models.
Rede de Atenção Psicossocial na cidade de Olinda: desenvolvimento de um protótipo web para melhoria do acesso aos serviços de saúde mental
(2024-10-04) Silva, Isaque João da; Andrade, Ermeson Carneiro de; Araújo, Carlos Julian Menezes; http://lattes.cnpq.br/3156174527107999; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/1877126987902999
This article highlights the importance of the Psychosocial Care Network (RAPS) in the mental health of the population and the need to provide clear and accessible information about its services in the city of Olinda. The objective of the study was to propose a web application, through the development of a prototype, that centralized essential information such as target audience, address, phone number, and operating hours of RAPS services. The methodology includes problem identiﬁcation, user proﬁle deﬁnition, ideation, prototype development, and evaluation through a questionnaire answered by 127 participants. The results show a positive evaluation of the prototype, with 91.4% of respondents agreeing on its navigability and 90.6% of participants expressing their intention to use it in their daily lives. It is concluded that the project is promising and reinforces the need for continuous improvements in the available information, as well as the possibility of implementing the proposed solution and adapting it to other municipalities.
Sentiment analysis of tweets related to SUS before and during COVID-19 pandemic
(2021-02-19) Silva, Henrique Farias Pereira da; Andrade, Ermeson Carneiro de; Araújo, Danilo Ricardo Barbosa de; Dantas, Jamilson Ramalho; http://lattes.cnpq.br/5655706091153128; http://lattes.cnpq.br/2466077615273972; http://lattes.cnpq.br/9810796504568932
The COVID-19 pandemic has affected the whole world since the beginning of 2020. In Brazil, over 70% of the population rely on the Brazil’s Unified Health System (SUS). Knowing public opinion related to SUS is very important for the improvement of services and assistance provided by such an entity. Sentiment analysis has been used in several applications including social networks and blogs to extract public opinion. Despite the fact that other papers have already worked with sentiment analysis, none of them have focused on SUS. Therefore, the goal of this paper is to analyse the sentiments shown by Brazillian Twitter users about SUS before and during COVID-19 pandemic. To reach this goal, a database of portuguese tweets regarding SUS posted between december 2019 and october 2020 was created. The tweets were pre-processed, classified and then analysed. The results show that, in most cases, users are in favor of SUS.