Relevance ranking metrics for learning objects

Xavier Ochoa, Erik Duval

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

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.

Original languageEnglish (US)
Title of host publicationCreating New Learning Experiences on a Global Scale - Second European Conference on Technology Enhanced Learning, EC-TEL 2007, Proceedings
PublisherSpringer Verlag
Pages262-276
Number of pages15
ISBN (Print)9783540751946
DOIs
StatePublished - 2007
Event2nd European Conference on Technology Enhanced Learning, EC-TEL 2007 - Crete, Greece
Duration: Sep 1 2007Sep 1 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4753 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd European Conference on Technology Enhanced Learning, EC-TEL 2007
Country/TerritoryGreece
CityCrete
Period9/1/079/1/07

Keywords

  • Learning objects
  • Personal relevance
  • Relevance ranking
  • Situational relevance
  • Topical relevance

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Relevance ranking metrics for learning objects'. Together they form a unique fingerprint.

Cite this