Software Architecture for Adaptation and Recommendation of Course Content and Activities Based on Learning Analytics

Nowadays the main challenge in learning analytics is to suggest efficient methods and technologies in order to achieve better learner results. This paper presents a software architecture for adaptation and recommendation of course content and activities based on learning analytics. It is comprised of layers for ingestion layer, aggregation layer, storage layer and big data processing and analyses layer. An algorithm for prediction of student learning based on machine learning for processing and analysis of data and knowledge discovery with respect to main learner and teacher activities is presented. The proposed algorithm for student learning classification is implemented by using Averaged Perceptron method. Experimental results are presented and discussed. The purpose of the study is to apply the software architecture on learning analytics by practical experiments for specific case study identifying event elements in sequenced learners' and courses' activities logs, and student learning prediction.