Resumo

CAMPOS WRIGHT, Wisleidys  e  TRUJILLO CASANOLA, Yaimí. Artificial Neural Networks in effort estimation. Rev cuba cienc informat [online]. 2021, vol.15, n.2, pp. 183-198.  Epub 01-Jun-2021. ISSN 2227-1899.

Planning is a critical part of project management work, that requires estimates of effort for a given project. Given the importance of meeting delivery deadlines while maintaining quality levels, the imperative to monitor and control the evolution of projects and the uncertainty generated by estimation, the need to create methods to solve these issues has arisen, which has aroused the interest of companies dedicated to software production. Researchers have developed machine learning algorithms, which allow a more accurate prediction of the effort to adjust the planning. Recently, techniques have been defined in the software industry, where artificial intelligence and algorithmic models are combined for effort estimation. This article presents the state of the art in the use of artificial neural networks for this purpose. A compilation of academic papers where neural networks are combined with hybrid algorithms based on the behavior of animals and insects for network learning was carried out, demonstrating the trend of its utilization to optimize effort estimation during early planning in software development projects.

Palavras-chave : software development; effort estimation; neural networks; hybrid algorithms.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )