PRIVACY PRESERVING LEARNING ANALYTICS

Abstract

Educational information contains valuable information which will be harvested through learning analytics to supply new insights for an improved education system. However, sharing or analysis of this data introduces privacy risks for the information subjects, largely students. Existing add the educational analytics literature identifies the requirement for privacy and cause attention-grabbing analysis instructions, however fails to use state of the art privacy protection ways with quantitative and mathematically rigorous privacy guarantees. This work aims to use and appraise such ways on learning analytics by approaching the matter from two perspectives: the information is named and so shared with a learning analytics expert, and also the learning analytics expert is given a privacy-preserving interface that governs her access to the information. We tend to develop proof-of-concept implementations of privacy conserving learning analytics tasks victimization each views and run them on real and artificial datasets. We tend to conjointly present an experimental study on the trade-off between individuals’ privacy and also the accuracy of the educational analytics tasks.