Gustavo de Assis Costa

Dr. Gustavo de Assis Costa

Computer Science, Instituto Federal de Goiás, Goiânia, Goiás, Brazil
orcid

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 August 2021 in Sensors
Verified
Journal Article
Open access
PDF file
NDNarjes Davari
Narjes Davari
BVBruno Miguel Veloso
Bruno Miguel Veloso
Gustavo de Assis CostaGustavo de Assis Costa
Gustavo de Assis Costa
PPPedro Mota Pereira
Pedro Mota Pereira
Rita P. RibeiroRita P. Ribeiro
Rita P. Ribeiro
João GamaJoão Gama
João Gama
In the last few years, many works have addressed Predictive Maintenance (PdM) by the use of Machine Learning (ML) and Deep Learning (DL) solutions, especially the latter. The monitoring and logging of industrial equipment events, like temporal behavior and fault events—anomaly detection in time-series—can be obtained from records generated by sensors installed in different parts of an industrial plant. However, such progress is incipient because we still have many challenges, and the performance of applications depends on the appropriate choice of the method. This article presents a survey of existing ML and DL techniques for handling PdM in the railway industry. This survey discusses the main approaches for this specific application within a taxonomy defined by the type of task, employed methods, metrics of evaluation, the specific equipment or process, and datasets. Lastly, we conclude and outline some suggestions for future research.

16 June 2025 in Smart Agricultural Technology
Verified
Journal Article
Open access
DGDanillo Gontijo
Danillo Gontijo
DSDouglas Rolins Santana
Douglas Rolins Santana
Gustavo de Assis CostaGustavo de Assis Costa
Gustavo de Assis Costa
Victor E. CabreraVictor E. Cabrera
Victor E. Cabrera
EFEduardo Noronha de Andrade Freitas
Eduardo Noronha de Andrade Freitas

1 March 2016
Verified
Conference Paper
Closed access
Gustavo de Assis CostaGustavo de Assis Costa
Gustavo de Assis Costa
JOJose Maria Parente De Oliveira
Jose Maria Parente De Oliveira
The amount and diversity of data in the Semantic Web has grown quite. RDF datasets has proportionally more problems than relational datasets due to the way data are published, usually without formal criteria. Entity Resolution is n important issue which is related to a known task of many research communities and it aims at finding all representations that refer to the same entity in different datasets. Yet, it is still an open problem. Blocking methods are used to avoid the quadratic complexity of the brute force approach by clustering entities into blocks and limiting the evaluation of entity specifications to entity pairs within blocks. In the last years only a few blocking methods were conceived to deal with RDF data and novel blocking techniques are required for dealing with noisy and heterogeneous data in the Web of Data. In this paper we present a blocking scheme, CER-Blocking, which is based on an inverted index structure and that uses different data evidences from a triple, aiming to maximize its effectiveness. To overcome the problems of data quality or even the very absence thereof, we use two blocking key definitions. This scheme is part of an ER approach which is based on a relational learning algorithm that addresses the problem by statistical approximation. It was empirically evaluated on real and synthetic datasets which are part of consolidated benchmarks found on the literature.

6 April 2024 in Advances in Computational Intelligence
Verified
Journal Article
Closed access
PDF file
DCde Andrade Cruvinel
de Andrade Cruvinel
LELucas Elias
Lucas Elias
WJWanderlei Malaquias Pereira Junior
Wanderlei Malaquias Pereira Junior
DCde Campos
de Campos
AIAmanda Isabela
Amanda Isabela
RDRoberto Viegas Dutra
Roberto Viegas Dutra

29 September 2025
Verified
Conference Paper
Open access
PDF file
JSJoão Vitor Felipe dos Santos
João Vitor Felipe dos Santos
RNRicardo Marçal de Andrade Nascimento
Ricardo Marçal de Andrade Nascimento
ACAdriano César de Melo Camargo
Adriano César de Melo Camargo
SCSergio Daniel Carvalho Canuto
Sergio Daniel Carvalho Canuto
Gustavo de Assis CostaGustavo de Assis Costa
Gustavo de Assis Costa
DSDaniel Xavier de Sousa
Daniel Xavier de Sousa
Este trabalho apresenta uma nova base de dados para Sistemas de Recomendação de Artigos Científicos (SRAC). Além de escassas, muitas das bases de dados no contexto SRAC utilizam apenas relações de coautoria como critério de relevância, negligenciando a importância das avaliações explícitas feitas por usuários. Para mitigar esse problema, propomos uma nova base de dados com rótulos de relevância explicitamente definidos, composta por mais de 2 mil pesquisadores, 30 áreas do conhecimento e aproximadamente 71 mil trabalhos associados. O trabalho inclui uma caracterização e avaliação da base proposta, juntamente com outras bases amplamente utilizadas na literatura.

4 August 2025 in Revista de Gestão Social e Ambiental
Verified
Journal Article
Open access
PDF file
JNJordana Portilho Neves
Jordana Portilho Neves
ACAndré Luis Brasil Cavalcante
André Luis Brasil Cavalcante
Gustavo de Assis CostaGustavo de Assis Costa
Gustavo de Assis Costa
TMTHIAGO AUGUSTO MENDES
THIAGO AUGUSTO MENDES
Objectives: This study aims to predict slope stability using machine learning algorithms, considering geotechnical and geometric variables to improve the accuracy of safety factor estimation and mitigate risks associated with slope failures. Theoretical Background: Traditional limit equilibrium methods, such as Bishop's method, present limitations when dealing with complex environmental conditions. Recent literature has highlighted the potential of machine learning techniques, such as Random Forest, Neural Networks, and XGBoost, in modeling nonlinear and highly variable systems, which are typical in geotechnical problems. Method: A parametric dataset with 1,520 observations generated through GeoSlope simulations was used, containing variables such as cohesion, internal friction angle, height, slope angle, and water level. Four algorithms were implemented: Random Forest, Gradient Boosting, Artificial Neural Networks, and XGBoost. Model performance was evaluated through cross-validation, hyperparameter tuning, and metrics including R², MSE, and RMSE. Results and Discussion: The XGBoost model achieved the best performance (R² = 0.9841; RMSE = 0.1419), followed by Gradient Boosting and Neural Networks. Variable importance analysis indicated that cohesion and internal friction angle are the primary factors influencing slope stability. Research Implications: The findings reinforce the applicability of machine learning models as decision-support tools in geotechnical engineering projects, offering greater precision and efficiency compared to traditional methods. Originality/Value: This study advances the field by integrating multiple modern algorithms within a replicable modeling framework, demonstrating superior performance over previous approaches and expanding the scope for practical applications in geotechnical engineering.