Requirement Change Prediction Model for Small Software Systems

The software industry plays a vital role in driving technological advancements. Software projects are complex and consist of many components, so change is unavoidable these projects. requirements must be predicted early to preserve resources, since it can lead project failures. This work focuses on small-scale systems which changed gradually. provides probabilistic prediction model, predicts the probability changes requirement specifications. first part considers analyzing due certain variables with help stakeholders, developers, experts by questionnaire method. Then, proposed model incorporates their knowledge Bayesian network as conditional probabilities independent dependent variables. approach utilizes variable elimination method obtain posterior revisions document. was evaluated sensitivity analysis comparison methods. For given dataset, computed low state 0.42, high 0.45. Thus, results proved that predict document accurately outperforming existing models.