The more Product Complexity, the more Actual Effort? An Empirical Investigation into Software Developments

摘要: [Background:] Software effort prediction methods and models typically assume positive correlation between software product complexity development effort. However, conflicting observations, i.e. negative actual effort, have been witnessed from our experience with the COCOMO81 dataset. [Aim:] Given doubt about whether observed phenomenon is a coincidence, this study tries to investigate if an increase in can result abovementioned counter-intuitive trend projects. [Method:] A modified association rule mining approach applied transformed To reduce noise of analysis, uses constant antecedent (Complexity increases while Effort decreases) mine potential consequents pruning. [Results:] The experiment has respectively mined four, five, seven rules general, embedded, organic projects data. suggested two main aspects, namely human capability scale, be particularly concerned study. [Conclusions:] not coincidence under particular conditions. In project, interactions other factors, such as Programmer Capability Analyst Capability, inevitably play "friction" role weakening practical influences on