Relevance Ranking Metrics for Learning Objects

摘要

Technologies that solve the scarce availability of learning objects have created the opposite problem: abundance of choice. The solution to that problem is relevance ranking. Unfortunately current techniques used to rank learning objects are not able to present the user with a meaningful ordering of the result list. This work interpret the Information Retrieval concept of Relevance in the context of learning object search and use that interpretation to propose a set of metrics to estimate the Topical, Personal and Situational relevance. These metrics are calculated mainly from usage and contextual information. An exploratory evaluation of the metrics shows that even the simplest ones provide statistically significant improvement in the ranking order over the most common algorithmic relevance metric.