The aim of the paper is to describe the design of a Web 3.0-based Intelligent Learning System (ILS) that addressing the students' needs in the 21st century. The design is based theoretically, on the principles of the connectivism theory and technically, it implements the semantic web representations combining with the use of learning analytics techniques. The work emphasises that implementing a learning analytics approach that uses: text classification, sentiment analysis, topics extraction, and text clustering on the basis of a semantic web and ontologies can support the connectivist learning. The semantic learning analytics process, represents the key element of the proposed intelligent learning analytics system to infer and deduce hidden data in the massive learning data thanks to semantic models of i-SoLearn. The aim is to guide students to understand through recommendations, charts and visualisations their learning behaviour and to give teachers feedbacks, enabling them to examine both students' learning and activities. An experimental study using i-SoLearn (an intelligent social learning environment), indicates that designing an ILS based on Web 3.0 techniques is effective and expected to show a great advantage in enhancing the connectivist learning of students in the digital age.

computers and education, semantic Web, technology enhanced learning, personalization and profiling, knowledge acquisition, artificial intelligence, content analysis and indexing