An Experimental Study on Collaborative Filtering for Requirements Engineering

Requirements elicitation and management are fundamental phases in every project, but difficulties in the interactions between the stakeholders can make them hard. Techniques which can reduce information and communication overloading could help to avoid misunderstandings. Recent works underline that Recommender Systems represent a useful technology that can ease the Requirements Engineering activity. This study presents an original recommender system (built on graph-theoretical concepts using combinatorial optimization methods) that can be employed for the problem of the requirements management. This work validates its proposal against a real case study.