PREDICCIÓN Y CLASIFICACIÓN DE RIESGOS EN PROYECTOS DE SISTEMAS DE INFORMACIÓN

Abstract

The development of a software project is subjected to many risks and treatments of diverse nature. The omission of risk management causes out of date product deliveries, with higher cost than planned and/or presenting failures concerning the specifications and requirements established by the client. It is considered of great utility for the Project Manager to find a model that, taking basic information of the project, would be able to predict its difficulty and to classify it depending on its risk, so that it would be able to propose coherent as well as proportional actions up to the level of problems expected during the development. The main objective of this work is to study the availability of the application of data mining techniques that allows to characterize the risk of software projects and to predict the approximate number and type of failures (extreme, major, minor) before the development phase. It will also identify the characteristic needed to develop the estimation, as project characteristics, developing language, use of CASE tools, team size, number of systems to integrate, using real data gathered by ISBG. In order to achieve this, the current systems are analyzed, selecting the more available characteristics in ISBG data base. Keywords: Software projects, Risk management, Data mining, Project Management