Previous researches on virtual collaborative communities emphasize more on the interactive behavior but ignore the collaborative behavior. To address this issue, this paper focuses on how to apply the tags which express the technical need of projects and the interest of cooperators to cooperator recommendation. This paper proposes a method of measuring the tags similarity from both text semantics perspective and relational semantics perspective based on items and cooperators, and constructs the work preference of cooperator in virtual collaborative communities from a new way. By defining a cooperator recommendation model, this paper also introduces the method of measuring the similarity of work preferences between cooperators and calculating the matching degree between cooperators and projects. The acquired information can then be used in cooperator recommendation algorithm. Furthermore, this paper investigates the popularity of tags and proposes the method of recommending project to community newcomer using tag popularity. Finally, by using the open-source community data from www.codeplex.com and comparing with other algorithm, the behavior of the proposed recommendation algorithm is verified. Results show that it gets a good recommendation effect in virtual collaborative communities and solves the problem of sparse matrix and cold start.
展开▼
