A research analytics framework-supported recommendation approach for supervisor selection

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalNot applicablepeer-review

3 Scopus Citations
View graph of relations


  • Mingyu ZHANG
  • Jian MA
  • Zhiying LIU
  • Jianshan SUN

Related Research Unit(s)


Original languageEnglish
Pages (from-to)1 - 18
Journal / PublicationBritish Journal of Educational Technology
Early online date9 Mar 2015
Publication statusPublished - Mar 2016


Identifying a suitable supervisor for a new research student is vitally important for his or her academic career. Current information overload and information disorientation have posed significant challenges for new students. Existing research for supervisor identification focuses on quality assessment of candidates, but ignores indirect relevance with candidate supervisors’ previous students, social network connections and their thinking styles. This paper presents a comprehensive student-centric approach based on research analytics framework for finding and recommending supervisors for new students. In particular, it integrates multiple measurements from three dimensions, ie, relevance, connectivity and quality. A prototype system was developed to support student–supervisor recommendations on a research social network platform (ie, www.ScholarMate.com). The results of user-based evaluations demonstrate that our proposed approach generates more satisfactory recommendations as compared with that of all baseline methods.