Relevance Ranking Metrics for Learning Objects

Member, IEEE Computer Society

Research output: Contribution to journalArticle

Abstract

The main objective of this paper is to improve the current status of learning object search. First, the current situation is analyzed and a theoretical solution, based on relevance ranking, is proposed. To implement this solution, this paper develops the concept of relevance in the context of learning object search. Based on this concept, it proposes a set of metrics to estimate the topical, personal, and situational relevance dimensions. These metrics are calculated mainly from usage and contextual information and do not require any explicit information from users. 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. Moreover, combining the metrics through learning algorithms sorts the result list 50 percent better than the baseline ranking.

Original languageEnglish (US)
Pages (from-to)34-48
Number of pages15
JournalIEEE Transactions on Learning Technologies
Volume1
Issue number1
DOIs
StatePublished - Jan 1 2008

Fingerprint

ranking
Learning algorithms
learning
evaluation

Keywords

  • learning object repository
  • Learning objects
  • metadata
  • RankNet
  • relevance ranking

ASJC Scopus subject areas

  • Education
  • Engineering(all)
  • Computer Science Applications

Cite this

Relevance Ranking Metrics for Learning Objects. / Member, IEEE Computer Society.

In: IEEE Transactions on Learning Technologies, Vol. 1, No. 1, 01.01.2008, p. 34-48.

Research output: Contribution to journalArticle

Member, IEEE Computer Society. / Relevance Ranking Metrics for Learning Objects. In: IEEE Transactions on Learning Technologies. 2008 ; Vol. 1, No. 1. pp. 34-48.
@article{8c63a4f6c8404753912be8a55ef6168a,
title = "Relevance Ranking Metrics for Learning Objects",
abstract = "The main objective of this paper is to improve the current status of learning object search. First, the current situation is analyzed and a theoretical solution, based on relevance ranking, is proposed. To implement this solution, this paper develops the concept of relevance in the context of learning object search. Based on this concept, it proposes a set of metrics to estimate the topical, personal, and situational relevance dimensions. These metrics are calculated mainly from usage and contextual information and do not require any explicit information from users. 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. Moreover, combining the metrics through learning algorithms sorts the result list 50 percent better than the baseline ranking.",
keywords = "learning object repository, Learning objects, metadata, RankNet, relevance ranking",
author = "{Member, IEEE Computer Society} and Xavier Ochoa and Erik Duval",
year = "2008",
month = "1",
day = "1",
doi = "10.1109/TLT.2008.1",
language = "English (US)",
volume = "1",
pages = "34--48",
journal = "IEEE Transactions on Learning Technologies",
issn = "1939-1382",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Relevance Ranking Metrics for Learning Objects

AU - Member, IEEE Computer Society

AU - Ochoa, Xavier

AU - Duval, Erik

PY - 2008/1/1

Y1 - 2008/1/1

N2 - The main objective of this paper is to improve the current status of learning object search. First, the current situation is analyzed and a theoretical solution, based on relevance ranking, is proposed. To implement this solution, this paper develops the concept of relevance in the context of learning object search. Based on this concept, it proposes a set of metrics to estimate the topical, personal, and situational relevance dimensions. These metrics are calculated mainly from usage and contextual information and do not require any explicit information from users. 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. Moreover, combining the metrics through learning algorithms sorts the result list 50 percent better than the baseline ranking.

AB - The main objective of this paper is to improve the current status of learning object search. First, the current situation is analyzed and a theoretical solution, based on relevance ranking, is proposed. To implement this solution, this paper develops the concept of relevance in the context of learning object search. Based on this concept, it proposes a set of metrics to estimate the topical, personal, and situational relevance dimensions. These metrics are calculated mainly from usage and contextual information and do not require any explicit information from users. 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. Moreover, combining the metrics through learning algorithms sorts the result list 50 percent better than the baseline ranking.

KW - learning object repository

KW - Learning objects

KW - metadata

KW - RankNet

KW - relevance ranking

UR - http://www.scopus.com/inward/record.url?scp=70350130350&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70350130350&partnerID=8YFLogxK

U2 - 10.1109/TLT.2008.1

DO - 10.1109/TLT.2008.1

M3 - Article

VL - 1

SP - 34

EP - 48

JO - IEEE Transactions on Learning Technologies

JF - IEEE Transactions on Learning Technologies

SN - 1939-1382

IS - 1

ER -