Extracting Prerequisite Relations Among Wikipedia Concepts Using the Clickstream Data.

A prerequisite relation describes a basic dependency relation between concepts in education, cognition and other fields. Especially, prerequisite relations among concepts play a very important role in various intelligent education applications, such as concept map extraction, learning object sequencing, reading order list generation. In this paper, we investigate the problem of extracting prerequisite relations among Wikipedia concepts. We take advantage of Wikipedia clickstream data and related concept sets to discover prerequisite relations among Wikipedia concepts. Evaluations on two datasets that include nine domains show that the proposed method can cover most of the concept pairs, and achieves significant improvements (+1.7–31.0% by Accuracy) comparing with existing methods.