A Personalized Explainable Learner Implicit Friend Recommendation Method

Abstract With the rapid development of social networks, academic networks have attracted increasing attention. In particular, providing personalized recommendations for learners considering data sparseness and cold-start scenarios is a challenging task. An important research topic to accurately discover potential friends build implicit learning groups obtain collaborative similar according content. This paper proposes explainable learner friend recommendation method (PELIRM). Methodologically, PELIRM utilizes learner's multidimensional interaction behavior in calculate degrees trust between applies three-degree influence theory mine learners. The similarity interests calculated by cosine term frequency–inverse document frequency. To solve problem learners, common check-in IP used location information. Finally, degree trust, interests, geographic distance are combined as ranking indicators recommend give multiple interpretations results. By verifying evaluating proposed on real from Scholar.com, experimental results show that reliable effective terms explainability.