Abstract

PEREZ VERA, Yasiel  and  BERMUDEZ PENA, Anié. Project stakeholder classification based on soft computing techniques. Rev cuba cienc informat [online]. 2018, vol.12, n.4, pp. 140-155. ISSN 2227-1899.

Stakeholder classification process is usually carried out by project manager using methods such as interviewing experts, brainstorming and checklists. These methods are carried out manually and subjectively by specialists belonging to the project. This affects the classification accuracy and project managers do not have more detailed information when making decisions about stakeholders. The objective of this research is to propose a genetic fuzzy system for classifying stakeholders for improving the classification quality with respect to the manually performed in projects. The proposal realizes the machine learning and adjustment of fuzzy inference systems for the stakeholder’s classification from the execution of six genetic algorithms: GFS.THRIFT, GFS.FR.MOGUL, GFS.GCCL, FH.GBML, GFS.LT.RS and SLAVE. It examines the results of applying them in 10 iterations by calculating the measures: accuracy, false positive, false negative, mean square error and symmetric mean absolute percentage error. The best results are shown by FH.GBML algorithm. The genetic fuzzy system implemented improves the stakeholder’s classification as a tool to support decision making in organizations oriented to production by projects.

Keywords : fuzzy inference system; genetic algorithm; project management; soft computing; stakeholder classification.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )