A grey rough set model for evaluation and selection of software cost estimation methods
Purpose – Due to the limitation of acknowledgment, the complexity of software system and the interference of noises, this paper aims to solve the traditional problem: traditional software cost estimation methods face the challenge of poor and uncertain inputs. Design/methodology/approach – Under such circumstances, different cost estimation methods vary greatly on estimation accuracy and effectiveness. Therefore, it is crucial to perform evaluation and selection on estimation methods against a poor information database. This paper presents a grey rough set model by introducing grey system theory into rough set based analysis, aiming for a better choice of software cost estimation method on accuracy and effectiveness. Findings – The results are very encouraging in the sense of comparison among four machine learning techniques and thus indicate it an effective approach to evaluate software cost estimation method where insufficient information is provided. Practical implications – Based on the grey rough set model, the decision targets can be classified approximately. Furthermore, the grey of information and the limitation of cognition can be overcome during the use of the grey rough interval correlation cluster method. Originality/value – This paper proposed the grey rough set model combining grey system theory with rough set for software cost estimation method evaluation and selection.