Navegando por Autor "Meneghini, Ivan Reinaldo"

Agora exibindo 1 - 3 de 3

An adaptive multi-objective algorithm based on decomposition and large neighborhood search for a green machine scheduling problem.
(2019) Cota, Luciano Perdigão; Guimarães, Frederico Gadelha; Ribeiro, Roberto Gomes; Meneghini, Ivan Reinaldo; Oliveira, Fernando Bernardes de; Souza, Marcone Jamilson Freitas; Siarry, Patrick
Green machine scheduling consists in the allocation of jobs in order to maximize production, in view of the sustainable use of energy. This work addresses the unrelated parallel machine scheduling problem with setup times, with the minimization of the makespan and the total energy consumption. The latter takes into account the power consumption of each machine in different operation modes. We propose multi-objective extensions of the Adaptive Large Neighborhood Search (ALNS) metaheuristic with Learning Automata (LA) to improve the search process and to solve the large scale instances efficiently. ALNS combines ad-hoc destroy and repair (also named removal and insertion) operators and a local search procedure. The LA is used to adapt the selection of insertion and removal operators within the framework of ALNS. Two new algorithms are developed: the MO-ALNS and the MO-ALNS/D. The first algorithm is a direct extension of single objective ALNS by using multi-objective local search. As this method does not offer much control of the diversification of the Pareto front approximation, a second strategy employs the decomposition approach similar to MOEA/D algorithm. The results show that the MO-ALNS/D algorithm has better performance than MO-ALNS and MOEA/D in all indicators. These findings show that the decomposition strategy is beneficial not only for evolutionary algorithms, but it is indeed an efficient way to extend ALNS to multi-objective problems.
ChurNAS : uma busca de arquitetura neural para previsão de desligamento de clientes.
(2023) Almeida, Marcus Daniel de; Moreira, Gladston Juliano Prates; Luz, Eduardo José da Silva; Moreira, Gladston Juliano Prates; Luz, Eduardo José da Silva; Silva, Ivair Ramos; Meneghini, Ivan Reinaldo
A Predição de desligamento de clientes PDC (do inglês Customer Churn Pre- diction) é fundamental para a gestão eficiente de clientes, uma vez que permite a otimização da lucratividade por meio de estratégias de marketing informadas e cam- panhas de retenção. Nesse contexto, o presente estudo propõe uma nova abordagem, chamada ChurNAS, baseada em algoritmo genético para a busca de arquiteturas neurais (Neural Architecture Search - NAS) em problemas de PDC na indústria de serviços financeiros. Ao contrário dos modelos tradicionais, como regressão logística e árvore de decisão, as redes neurais profundas apresentam maior versatilidade para modelagem de dados complexos. No entanto, a busca pela arquitetura ideal em redes neurais profundas ́e um desafio devido `a sua alta complexidade. Os resultados demonstram que a abordagem ChurNAS encontrou modelos com desempenho supe- rior aos classificadores tradicionais ajustados por otimiza ̧c ̃ao de hiperparâmetros. A abordagem proposta obteve uma acurácia de 88,6%, em comparação com 82,54% do XG-Boost e 82,49% do Floresta Aleatória. Al ́em disso, alcançou uma sensibilidade de 58,89%, enquanto o XG-Boost e o Floresta Aleatória apresentaram 57,1% e 57,81%, respectivamente. Quanto `a precisão, a abordagem ChurNAS obteve 39,41%, superando o XG-Boost (26,96%) e o Floresta Aleatória (26,17%). Adicionalmente, o estudo examinou o impacto da quantidade de dados e da capacidade do modelo, enfatizando a importância de considerar a natureza temporal das transações financei- ras ao utilizar redes neurais para PDC. Em suma, este trabalho destaca o potencial da abordagem ChurNAS para solucionar problemas de PDC no setor de servi ̧cos financeiros e melhorar a eficiência do gerenciamento de clientes.
Self-supervised learning for arrhythmia classification.
(2023) Silva, Guilherme Augusto Lopes; Luz, Eduardo José da Silva; Silva, Pedro Henrique Lopes; Luz, Eduardo José da Silva; Silva, Pedro Henrique Lopes; Freitas, Vander Luis de Souza; Meneghini, Ivan Reinaldo
Arrhythmias, heart diseases that are commonly diagnosed through electrocar- diograms (ECG), require computational methods for detection and classification to improve the physician’s diagnosis. Although there is abundant literature on the subject, the high intra-patient variability and noise of ECG signals pose challenges in developing practical machine-learning models. To address this, we propose a cus- tomized adjustment of machine learning models through self-supervised learning with human-in-the-loop. Our approach introduces a pretext task called ECGWavePuzzle, which improves classification performance through better generalization. Evaluation metrics on the MIT-BIH database demonstrate the effectiveness of our approach, which improved the ECGnet global accuracy by over 10% and the Mousavi’s CNN by over 13%. Additionally, the experimental results demonstrated that the proposed approach improved the sensitivity and positive predictive value of the arrhythmic classes for certain patients.