Rodrigo Francisquini da Silva

Idiomas

Produções bibliográficas

• FRANCISQUINI, RODRIGO ; LORENA, ANA CAROLINA ; NASCIMENTO, MARIÁ C.V. . Community-based anomaly detection using spectral graph filtering. APPLIED SOFT COMPUTING , v. 118, p. 108489, 2022.

• DA SILVA, TIAGO TIBURCIO ; FRANCISQUINI, RODRIGO ; NASCIMENTO, MARIÁ C.V. . Meteorological and human mobility data on predicting COVID-19 cases by a novel hybrid decomposition method with anomaly detection analysis: A case study in the capitals of Brazil. EXPERT SYSTEMS WITH APPLICATIONS , v. 182, p. 115190, 2021.

• FRANCISQUINI, RODRIGO ; BERTON, RAFAEL ; SOARES, SANDRO GOMES ; PESSOTTI, DAYELLE S. ; CAMACHO, MAURÍCIO F. ; ANDRADE-SILVA, DÉBORA ; BARCICK, UILLA ; SERRANO, SOLANGE M.T. ; CHAMMAS, ROGER ; NASCIMENTO, MARIÁ C.V. ; ZELANIS, ANDRÉ . Community-based network analyses reveal emerging connectivity patterns of protein-protein interactions in murine melanoma secretome. Journal of Proteomics , v. 232, p. 104063, 2021.

• FRANCISQUINI, RODRIGO ; ROSSET, VALÉRIO ; NASCIMENTO, MARIÁ C.V. . GA-LP: A genetic algorithm based on Label Propagation to detect communities in directed networks. Expert Systems with Applications , v. 74, p. 127-138, 2017.

• FRANCISQUINI, RODRIGO ; DA SILVA, TIAGO T. ; NASCIMENTO, MARIA C. V. . Detecting Anomalies In Daily COVID-19 Cases Data From Brazil Capitals Using GSP Theory. In: 2021 IEEE Congress on Evolutionary Computation (CEC), 2021, Kraków. 2021 IEEE Congress on Evolutionary Computation (CEC), 2021. p. 1296.

• FRANCISQUINI, RODRIGO ; NASCIMENTO, MARIA C. V. ; BASGALUPP, MARCIO P. . NGA-LP: A Robust and Improved Genetic Algorithm to Detect Communities in Directed Networks. In: 2018 IEEE Congress on Evolutionary Computation (CEC), 2018, Rio de Janeiro. 2018 IEEE Congress on Evolutionary Computation (CEC), 2018. p. 1.

• FRANCISQUINI, R ; NASCIMENTO, MARIÁ C.V. ; ROSSET, VALÉRIO . Uma heurística espectral local para a detecção de comunidades em redes. In: XLVII Simpósio Brasileiro de Pesquisa Operacional, 2015, Porto de Galinhas. Anais do XLVII Simpósio Brasileiro de Pesquisa Operacional, 2015.

Projetos de pesquisa

• 2017 - Atual

  Análise espectral para detecção de anomalias em redes dinâmicas com atributos, Descrição: Data anomaly detection strategies have several applications, such as intrusion detection in computer networks or fraud detection in financial transactions. In particular, when the data are representable through graphs, metrics and methods based on graph theory and known for the quality of the solutions obtained can be used. In the case of static graphs, the anomaly detection problem has been extensively studied and several algorithms have been proposed. However, there are few works focused on the detection of anomalies in dynamic networks with attributes. While most strategies consider node and edge updates, none of these strategies consider the history of these changes during anomaly detection. In addition, few strategies are scalable to deal with Big Data. In this case, anomaly detection strategies that use clustering algorithms are considered a good option, as they allow the analysis of groups of vertices instead of individual vertices, and, therefore, have a lower computational cost. Thus, this project aims to investigate existing unsupervised methods for anomaly detection in dynamic networks with attributes. As a main contribution, a scalable anomaly detection strategy in dynamic networks with attributes will be proposed. This strategy will use a clustering algorithm and spectral operators that will also be developed in the project. The developed strategy will be subjected to tests to attest its efficiency and compare the results obtained with the results of the best strategies in the literature.. , Situação: Em andamento; Natureza: Pesquisa. , Integrantes: Rodrigo Francisquini da Silva - Integrante / Mariá Cristina Vasconcelos Nascimento Rosset - Coordenador.

• 2015 - 2016

  GALP: Um algoritmo genético baseado no Label Propagation para o problema de detecção de comunidades em redes., Descrição: Many real-world networks have a topological structure characterized by cohesive groups of vertices. To perform the task of identifying such subsets of vertices, the problem of detecting communities in networks has aroused the interest of researchers and professionals. Despite the existence of several efficient community detection algorithms in the literature, most of them use global information about the network, not being applicable to distributed or large-scale networks. Furthermore, one of the main challenges in detecting communities in directed networks is the interpretation of network arcs with asymmetric relationships between vertices. This Scientific Initiation project proposed a genetic algorithm based on Label Propagation to detect communities in targeted networks. In computational experiments, the proposed strategy showed an excellent performance, being fast and achieving the best results on average in the networks tested.. , Situação: Concluído; Natureza: Pesquisa. , Integrantes: Rodrigo Francisquini da Silva - Integrante / Mariá Cristina Vasconcelos Nascimento Rosset - Coordenador / ROSSET, VALÉRIO - Integrante.

• 2014 - 2015

  Uma heurística espectral para o problema de detecção de comunidades em redes., Descrição: Spectral heuristics for problems of clustering in graphs or detection of communities in networks have been the subject of constant study, mainly due to the advancement of numerical methods for the determination of the set of eigenvalues and eigenvectors of matrices. Allied to this, the mathematical elegance and the quality of the solutions obtained by these heuristics are factors of great appeal for their study. The main focus of the project was to acquire knowledge of the graph clustering problem, more specifically, the modularity maximization problem. We developed a heuristic capable of detecting communities in networks using spectral theory.. , Situação: Concluído; Natureza: Pesquisa. , Integrantes: Rodrigo Francisquini da Silva - Integrante / Mariá Cristina Vasconcelos Nascimento Rosset - Coordenador / ROSSET, VALÉRIO - Integrante.

• 2013 - 2014

  Otimização linear e estudo de medidas de risco., Descrição: In this project it was studied how to model real situations with optimization problems. We focused more on linear problems (LP), for which we studied their structure in detail and the Simplex method as a computational tool to solve them. An important application of optimization theory is risk minimization in financial investments. We studied the Markowitz risk models, the Value at Risk (VaR) and the Conditional Value at Risk (CVaR). A geometric approach to these problems was developed and the formulation of CVaR minimization as PL was studied and implemented.. , Situação: Concluído; Natureza: Pesquisa. , Integrantes: Rodrigo Francisquini da Silva - Integrante / Luis Felipe Cesar da Rocha Bueno - Coordenador.