Educational Data Mining for Peer Assessment in Communities of Learners

In the last years, the design and implementation of web-based education systems has grown exponentially, spurred by the fact that neither students nor teachers are bound to a specific location and that this form of computer-based education is virtually independent of any specific hardware platform. These systems accumulate a large amount of data: educational data mining and learning analytics are the two much related fields of research with the aim of using these educational data to improve the learning process. In this chapter, the authors investigate the peer assessment setting in communities of learners. Peer assessment is an effective didactic strategy, useful to evaluate groups of students in educational environments such as high schools and universities where students are required to answer open-ended questions to increase their problem-solving skills. Furthermore, such an approach could become necessary in the learning contexts where the number of students to evaluate could be very large as, for example, in massive open online courses. Here the author focus on the automated support to grading open answers via a peer evaluation-based approach, which is mediated by the (partial) grading work of the teacher, and produces a (partial as well) automated grading. The author propose to support such automated grading by means of two methods, coming from the data-mining field, such as Bayesian Networks and K-Nearest Neighbours (K-NN), presenting some experimental results, which support our choices.