Talent Circle Detection in Job Transition Networks
Huang Xu*, Northwestern Polytechnical Uni; Jingyuan Yang, Rutgers University; zhi wen Yu, ; Hui Xiong, Rutgers; Hengshu Zhu, Baidu Inc.
With the high mobility of talent, it becomes critical for the recruitment team to ﬁnd the right talent from the right source in an eﬃcient manner. The prevalence of Online Professional Networks (OPNs), such as LinkedIn, enables the new paradigm for talent recruitment and job search. How-ever, the dynamic and complex nature of such talent information imposes signiﬁcant challenges to identify prospective talent sources from large-scale professional networks. Therefore, in this paper, we propose to create a job transition network where vertices stand for organizations and a directed edge represents the talent ﬂow between two organizations for a time period. By analyzing this job transition network, it is able to extract talent circles in a way such that every circle includes the organizations with similar talent exchange patterns. Then, the characteristics of these talent circles can be used for talent recruitment and job search. To this end, we develop a talent circle detection model and design the corresponding learning method by maximizing the Normalized Discounted Cumulative Gain (NDCG) of inferred probability for the edge existence based on edge weights. Then, the identiﬁed circles will be labeled by the representative organizations as well as keywords in job descriptions. Moreover, based on these identiﬁed circles, we develop a talent exchange prediction method for talent recommendation. Finally, we have performed extensive experiments on real-world data. The results show that, our method can achieve much higher modularity when comparing to the benchmark approaches, as well as high precision and recall for talent exchange prediction.
Filed under: Graph Mining and Social Networks