SkillsRec: A Novel Semantic Analysis Driven Learner Skills Mining and Filtering Approach for Personal Learning Environments Based on Teacher Guidance

Abstract

This paper presents SkillsRec - a novel teacher guidance based learner skills mining and filtering approach that identifies learner skills for Personal Learning Environment (PLE) based learning scenarios using Latent Semantic Analysis (LSA) technique. Skills Rec is developed on PLE design and development principles of the guided PLEs model [1]. Skills Rec takes teacher competencies/roles [2] and learner interests as input, melds them using LSA, and returns learner skills for the PLE-based learning as output. We compare learner-skill similarity scores of the Skills Rec with those generated through conventional Information Retrieval (IR) and Keywords Matching (KM) techniques. The aim is to report Skills Rec gains over conventional IR techniques. Based on Skills Rec results, this paper also provides top N=8 user-user recommendations most likely to be similar for a given active learner as a testing data.