CAJ | 학술논문

为研究如何帮助社交网络中虚拟团队从海量的用户中筛选更多行为相似的用户加入团队，建立了一种基于社交网络的团队成员推荐模型，为虚拟团队推荐一些行为特征相似的用户作为候选团队成员。通过对用户信息和团队信息进行分词处理，得到用户标签向量集和团队标签向量集；将基于遗传算法的聚类算法应用于团队标签向量集，得到团队聚类簇和团队聚类中心；在此基础上，将团队聚类中心作为用户集的初始聚类中心，聚类出行为特征与团队聚类中心相似的用户簇，将这些用户簇作为相应团队聚类簇的候选推荐用户；再应用协同过滤思想，筛选出团队推荐成员列表。实验结果表明，该算法有效地解决了虚拟团队成员推荐问题。
위연구여하방조사교망락중허의단대종해량적용호중사선경다행위상사적용호가입단대，건립료일충기우사교망락적단대성원추천모형，위허의단대추천일사행위특정상사적용호작위후선단대성원。통과대용호신식화단대신식진행분사처리，득도용호표첨향량집화단대표첨향량집；장기우유전산법적취류산법응용우단대표첨향량집，득도단대취류족화단대취류중심；재차기출상，장단대취류중심작위용호집적초시취류중심，취류출행위특정여단대취류중심상사적용호족，장저사용호족작위상응단대취류족적후선추천용호；재응용협동과려사상，사선출단대추천성원렬표。실험결과표명，해산법유효지해결료허의단대성원추천문제。
In order to help some virtual teams select users with similar behaviors to expand the scale of the teams, a new model for team members recommendation in social network is proposed to rec-ommend the users with similar behavior characteristics as the candidate members for a virtual team Firstly, it deals with the user information and team information by using words segmentation algo-rithm so as to get user label vector set and team label vector set.Secondly, the team clusters and team cluster centers are obtained by applying the clustering algorithm based on genetic algorithm to the team label vector set.Thirdly, the team cluster centers are used as the initial cluster centers to get the user clusters having similar behaviors to team cluster centers, and the user clusters are the candidates for related team clusters.Finally, the list of recommended members in the team is ob-tained through collaborative filtering.The experimental results show that the algorithm can effectively solve the problem of virtual team member recommendation.