European countries have accumulated an enormous quantity of information in Digital Libraries (DLs). Offering seamless universal access to those collections will have a formidable impact on citizens' activities. Students could use information in DLs for improving their curricula, but it is difficult to find the exact chunk of material that solves a specific problem. A possible solution is to develop technologies that learn user preferences for customising information search. This paper focuses on a system based on Machine Learning techniques, the Profile Extractor, which automatically builds student models. An experimental session has been performed, evaluating the accuracy of the system.
Student profiles to improve searching in e-learning systems
SEMERARO, Giovanni
2007-01-01
Abstract
European countries have accumulated an enormous quantity of information in Digital Libraries (DLs). Offering seamless universal access to those collections will have a formidable impact on citizens' activities. Students could use information in DLs for improving their curricula, but it is difficult to find the exact chunk of material that solves a specific problem. A possible solution is to develop technologies that learn user preferences for customising information search. This paper focuses on a system based on Machine Learning techniques, the Profile Extractor, which automatically builds student models. An experimental session has been performed, evaluating the accuracy of the system.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
