中文摘要：

如何利用社会网络信息来寻找一个合作高效、高质量的团队,已成为热门的研究话题.但现有团队生成问题中对个体拥有技能的度量大多都采用0-1方式,而在实际应用中如何界定个体是否拥有该技能的方法会在很大程度上影响团队完成任务的效率.另外在目前的基于社会网络的团队生成方法研究中,计算个体间关系强度时只考虑个体间曾经合作任务的数目,并没有深入挖掘社会网络条件下个体间的社会关系类别以及个体自身的其他属性,这些因素很大程度上也会影响个体间的社会关系,进而影响个体间的团队合作.针对以上问题,该文首先给出团队生成问题的具体定义和相关概念,给出技能贡献度的定义,并利用社会网络个体间的关系类别和个体间对应社会属性相似度引入一种关系模型来进一步量化团队成员个体间的关系强度;然后根据团队的不同形式分别进行了无领导者团队生成方法的研究和有领导者团队生成方法的研究,并分别提出了MCSTFA算法（Minimum Covering Steiner-based Team Forming Algorithm）和MSCTFA算法（Minimum Set Covering-based Team Forming Algorithm）来寻找最佳无领导者团队以及提出MLDTFA算法（Minimum Leader Distance based Team Forming Algorithm）来寻找最佳领导者和最佳团队.最后,利用DBLP数据集设计和实现实验以验证上述所有方法的可行性和有效性,并从团队合作代价、团队成员数量、团队连通性以及社会网络影响因素对算法的影响对比结果等方面进行比较和分析,实验结果验证了文中所提算法的可行性和高效性.

英文摘要：

How to use social network information to find a team with efficient collaborating performance and good quality has become a hot research issue in social networks. Current researches usually use 0-1 value to measure whether an individual has a skill or not while in practical applica- tions the way to judge whether an individual has certain skills will largely influence the efficiency of the team to complete the task. Moreover, most present studies only focus on considering the number of the past collaborated tasks and do not deeply explore different social relationships between individual as well as the individua＇s own other properties while calculating the strength of relationships between individuals in social network. These factors immensely influence social relationships and team cooperation between individuals. To solve the above problems, firstly, we give the problem definition and the related concepts as well as the definition of skill contribution and make use of a relation model to calculate the relationship between individuals which takes into consideration the categories of social connections and the properties of individuals. Secondly, according to the different forms of team, we study two types of team formation problems, team formation problem without a leader and team formation problem with a leader. Then we propose Minimum Covering Steiner-based Team Forming Algorithm and Minimum Set Covering-based Team Forming Algorithm to search for the best team without a leader separately and propose Minimum Leader Distance based Team Forming Algorithm to search the best team with a leader. Finally, we use DBLP as dataset and conduct experiments to verify the feasibility and effectiveness of all of the above methods, and compare our methods with existing algorithms from the following aspects： team cooperation cost, the cardinality of the team, team connectivity and the comparison results of social influence factors to algorithms. The experimental results prove the feasibility and effectiveness of our method