The Effectiveness of Personalization in Delivering E-learning Classes

In this paper we present an evaluation methodology of personalized delivery of multimedia resources in an e-learning platform that takes advantage of Web 2.0 technologies and its standards, such as (i) streaming using RSS and Atom feeds, (ii) searching using meta-data search engine, (iii) delivering using ubiquitous computing and (iv) personalized learning. Our framework consists of the following phases: (1) tracking the preferred delivery mechanism for each student model, (2) measuring the effect of the semantic profile time window parameter, (3) clustering students models based on similarity metrics on the delivery preferences and (4) comparing the results between personalized and non-personalized delivery. One important aspect of our approach is the combination of personalization and the data driven extraction (via clustering) of the students models based on similarity metrics. Our experimental results show that personalized delivery increases the usage of e-learning materials and the percentage of reviewing multimedia resources if these materials are delivered in ways that fit students preferences.