Application or user-centered design methods and data mining to define recommendations, wich fosters the use of the forum in a virtual learning experience

Abstract

The use of recommendation systems in learning virtual environments is increasingly becoming a feasible approach for providing the adaptive support required to attend students’ learning needs. With the interaction data obtained from these virtual environments it is possible to find indicators where data mining and machine learning techniques can be applied to identify relevant information that allows for the definition of recommendations. In this research we have applied unsupervised learning techniques to identify common interaction patterns with available forums in a course on the OpenACS/dotLRN platform. This will allow recommendations to be defined that help improve the students’ learning experience.