Rebecca Salles

Rebecca Salles

Auto-generated profile, based on publicly available information

Rebecca Salles

INRIA

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
1 January 2019 in Knowledge-Based Systems
Verified
Journal Article
Closed access
RSRebecca Salles
Rebecca Salles
KBKele Belloze
Kele Belloze
FPFabio Porto
Fabio Porto
PGPedro H. Gonzalez
Pedro H. Gonzalez
EOEduardo Ogasawara
Eduardo Ogasawara

1 November 2016 in Ecological Informatics
Verified
Journal Article
Closed access
RSRebecca Salles
Rebecca Salles
PMPatricia Mattos
Patricia Mattos
AIAna-Maria Dubois Iorgulescu
Ana-Maria Dubois Iorgulescu
EBEduardo Bezerra
Eduardo Bezerra
LLLeonardo De Lima
Leonardo De Lima
EOEduardo Ogasawara
Eduardo Ogasawara

1 January 2022 in Neurocomputing
Verified
Journal Article
Closed access
RSRebecca Salles
Rebecca Salles
EPEsther Pacitti
Esther Pacitti
EBEduardo Bezerra
Eduardo Bezerra
FPFabio Porto
Fabio Porto
EOEduardo Ogasawara
Eduardo Ogasawara

28 September 2020
Verified
Conference Paper
Closed access
PDF file
RSRebecca Salles
Rebecca Salles
LELuciana Escobar
Luciana Escobar
LBLais Baroni
Lais Baroni
RZRoccio Zorrilla
Roccio Zorrilla
AZArtur Ziviani
Artur Ziviani
EOEduardo Ogasawara
Eduardo Ogasawara
Ao analisar séries temporais é possível observar mudanças significativas no comportamento das observações que frequentemente caracterizam a ocorrência de eventos. Eventos se apresentam como anomalias, pontos de mudança, ou padrões frequentes. Na literatura existem diversos métodos para detecção de eventos. Entretanto, a busca por um método adequado para uma série temporal não é uma tarefa simples, principalmente considerando-se que a natureza dos eventos muitas vezes não é conhecida. Neste contexto, este trabalho apresenta Harbinger, um framework para integração e análise de métodos de detecção de eventos. O Harbinger foi avaliado em dados sintéticos e reais, onde foi possível constatar que suas funcionalidades promovem a seleção de métodos e a compreensão dos eventos detectados.

1 May 2017
Verified
Conference Paper
Closed access
RSRebecca Salles
Rebecca Salles
LALaura Assis
Laura Assis
GGGustavo Guedes
Gustavo Guedes
EBEduardo Bezerra
Eduardo Bezerra
FPFabio Porto
Fabio Porto
EOEduardo Ogasawara
Eduardo Ogasawara
Time series prediction has been attracting interest of researchers due to its increasing importance in decision-making activities in many fields of knowledge. The demand for better accuracy in time series prediction furthered the arising of many machine learning time series prediction methods (MLM). Choosing a suitable method for a particular dataset is a challenge and demands established benchmark methods (BM) for performance assessment. Suppose a particular BM is selected, and an experimental comparison is made with a particular MLM. If the latter does not provide better prediction results for the same dataset, this indicates that some improvements are needed for the MLM. Regarding this matter, adopting a well-established, easy to interpret, and tuned BM is desirable. This paper presents a framework for systematic benchmarking some MLM against well-known Linear Methods (LM), namely Polynomial Regression and models in the ARIMA family, used as BM for univariate time series prediction. We implemented such a framework within the R-Package named TSPred. This implementation was evaluated using a wide number of datasets from past prediction competitions. The results show that fittest LM provided by TSPred are adequate BM for univariate time series predictions.

21 January 2025 in Information Systems
Verified
Journal Article
Closed access
RSRebecca Salles
Rebecca Salles
BLBenoit Lange
Benoit Lange
RAReza Akbarinia
Reza Akbarinia
FMFlorent Masseglia
Florent Masseglia
EOEduardo Ogasawara
Eduardo Ogasawara
EPEsther Pacitti
Esther Pacitti