HCHalcyon Davys Pereira De Carvalho

Halcyon Davys Pereira De Carvalho

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Halcyon Davys Pereira De Carvalho has a total of 5 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Ensembles for Regression: A Statistical Analysis and Recommendations

    by Halcyon Carvalho, Joao Fausto Oliveira, Roberta Fagundes

    Ensemble techniques have proven effective in improving the performance of regression models. However, the decision between homogeneous and heterogeneous strategies is often made empirically, without considering the statistical properties of the data. This paper presents a framework that conducts a statistical analysis of datasets to recommend the most appropriate ensemble strategy. The framework leverages metrics such as coefficient of variation, skewness, kurtosis, correlation, and outlier presence to classify datasets into homogeneous or heterogeneous profiles. Ten real-world datasets with varying levels of complexity were evaluated. The results show that the proposed framework achieved the best performance in 80% of the cases, confirming that the statistical structure of the data should guide the selection of ensemble strategies. Wilcoxon statistical tests further validated the significance of the results. The proposed framework offers a systematic alternative to guide ensemble decisions in regression tasks, enhancing accuracy and reducing reliance on empirical trial-and-error procedures.

  • Dynamic selection of ensemble-based regression models: Systematic literature review

    by Halcyon Davys P. de Carvalho, João Fausto L. de Oliveira, Roberta Andrade de A. Fagundes


  • Open Access

    A Comparison Study About Parameter Optimization Using Swarm Algorithms

    by Halcyon Davys Pereira De Carvalho, Wedson L. Soares, Wylliams Barbosa Santos, Roberta Fagundes

    Adjusting the parameters of a machine learning algorithm can be difficult if the possible domain of expansion of these parameters is too high. In addition, if a sensible parameter is not adjusted correctly, the changes can be very impactful in the final results, making adjusting it manually not trivial. In order to adjust these features automatically, the current work proposes six models based on the use of optimization algorithms to adjust the models’ parameters automatically. These models were built around two machine learning-based algorithms, an extreme learning machine neural network, and a support vector regression. The optimization algorithms used are Particle Swarm Optimization, the Artificial Bee Colony, and the genetic algorithm. The models were compared with each other based on predictive precision in the criterion of Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and statistical tests. The experimental results on ten datasets in different contexts indicated that optimized algorithms models perform better in convergence, precision, and robustness than the non-optimized algorithms models. Therefore, the automatic adjustment of the parameters of optimized algorithms is a powerful tool for analyzing different data contexts. Thus, this study shows that the optimized algorithm models (in particular the ELM PSO model) are more accurate than all experimental evaluations.


  • Open Access

    Extreme Learning Machine Applied to Software Development Effort Estimation

    by Halcyon Davys Pereira De Carvalho, Roberta Fagundes, Wylliams Santos

    The project management process has been used in the area of Software Engineering to support project managers to keep projects under control. One of the essential processes in Software Engineering is to conduct an accurate and reliable estimation of the required effort to complete the project. This article’s objectives are: i) to identify the variables that influence the estimation based on the correlation, and ii) to apply the Extreme Learning Machine - ELM model for effort estimation and compare it with the literature models. Thus, it was investigated which technique has better effort prediction accuracy. The models were compared with each other based on predictive precision in the criterion of absolute mean residue (MAR) and statistical tests. The main findings in this study were: i) important variables for effort estimation and; ii) the results indicated that the ELM model presents the best results compared to the models in the literature for estimating software design effort. In this way, the use of Machine Learning techniques in the effort estimation process can increase the chances of success in the accuracy of the time estimates and the project’s costs.