ILOG

A framework for automatic annotation of learning objects with empirical usage metadata

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Learning objects (LOs) are digital or non-digital entities used for learning, education or training commonly stored in repositories searchable by their associated metadata. Unfortunately, based on the current standards, such metadata is often missing or incorrectly entered making search difficult or impossible. In this paper, we investigate automating metadata generation for SCORM-complaint LOs based on user interactions with the LO and static information about LOs and users. Our framework, called the Intelligent Learning Object Guide (iLOG), involves real-time tracking of each user sessions (an LO Wrapper), offline data mining to identify key attributes or patterns on how the LOs have been used as well as characteristics of the users (MetaGen), and the selection of these findings as metadata. Mechanisms used in the data mining include data imputation via clustering, association rule mining, and feature selection ensemble. This paper describes the methodology of automatic annotation, presents the results on the evaluation and validation of the algorithms, and discusses the resulting metadata. We have deployed our iLOG implementation for five LOs in introductory computer science topics and collected data for over 1400 sessions. We demonstrate that iLOG successfully tracks user interactions that can be used to automate the generation of meaningful empirical usage metadata for different stakeholder groups including learners and instructors, LO developers, and researchers.

Original languageEnglish (US)
Pages (from-to)215-236
Number of pages22
JournalInternational Journal of Artificial Intelligence in Education
Volume21
Issue number3
DOIs
StatePublished - Jul 6 2012

Fingerprint

Metadata
learning
Data mining
Association rules
Computer science
Feature extraction
Education
interaction
complaint
computer science
instructor
stakeholder

Keywords

  • Association rule mining
  • Data imputation
  • Empirical usage metadata
  • Feature selection
  • SCORM learning objects

ASJC Scopus subject areas

  • Education
  • Computational Theory and Mathematics

Cite this

@article{1bb3665709ab4b35a13887852c3aa1df,
title = "ILOG: A framework for automatic annotation of learning objects with empirical usage metadata",
abstract = "Learning objects (LOs) are digital or non-digital entities used for learning, education or training commonly stored in repositories searchable by their associated metadata. Unfortunately, based on the current standards, such metadata is often missing or incorrectly entered making search difficult or impossible. In this paper, we investigate automating metadata generation for SCORM-complaint LOs based on user interactions with the LO and static information about LOs and users. Our framework, called the Intelligent Learning Object Guide (iLOG), involves real-time tracking of each user sessions (an LO Wrapper), offline data mining to identify key attributes or patterns on how the LOs have been used as well as characteristics of the users (MetaGen), and the selection of these findings as metadata. Mechanisms used in the data mining include data imputation via clustering, association rule mining, and feature selection ensemble. This paper describes the methodology of automatic annotation, presents the results on the evaluation and validation of the algorithms, and discusses the resulting metadata. We have deployed our iLOG implementation for five LOs in introductory computer science topics and collected data for over 1400 sessions. We demonstrate that iLOG successfully tracks user interactions that can be used to automate the generation of meaningful empirical usage metadata for different stakeholder groups including learners and instructors, LO developers, and researchers.",
keywords = "Association rule mining, Data imputation, Empirical usage metadata, Feature selection, SCORM learning objects",
author = "Miller, {L. D.} and Leen-Kiat Soh and Samal, {Ashok K} and Nugent, {Gwen C}",
year = "2012",
month = "7",
day = "6",
doi = "10.3233/JAI-2012-024",
language = "English (US)",
volume = "21",
pages = "215--236",
journal = "International Journal of Artificial Intelligence in Education",
issn = "1560-4292",
publisher = "IOS Press",
number = "3",

}

TY - JOUR

T1 - ILOG

T2 - A framework for automatic annotation of learning objects with empirical usage metadata

AU - Miller, L. D.

AU - Soh, Leen-Kiat

AU - Samal, Ashok K

AU - Nugent, Gwen C

PY - 2012/7/6

Y1 - 2012/7/6

N2 - Learning objects (LOs) are digital or non-digital entities used for learning, education or training commonly stored in repositories searchable by their associated metadata. Unfortunately, based on the current standards, such metadata is often missing or incorrectly entered making search difficult or impossible. In this paper, we investigate automating metadata generation for SCORM-complaint LOs based on user interactions with the LO and static information about LOs and users. Our framework, called the Intelligent Learning Object Guide (iLOG), involves real-time tracking of each user sessions (an LO Wrapper), offline data mining to identify key attributes or patterns on how the LOs have been used as well as characteristics of the users (MetaGen), and the selection of these findings as metadata. Mechanisms used in the data mining include data imputation via clustering, association rule mining, and feature selection ensemble. This paper describes the methodology of automatic annotation, presents the results on the evaluation and validation of the algorithms, and discusses the resulting metadata. We have deployed our iLOG implementation for five LOs in introductory computer science topics and collected data for over 1400 sessions. We demonstrate that iLOG successfully tracks user interactions that can be used to automate the generation of meaningful empirical usage metadata for different stakeholder groups including learners and instructors, LO developers, and researchers.

AB - Learning objects (LOs) are digital or non-digital entities used for learning, education or training commonly stored in repositories searchable by their associated metadata. Unfortunately, based on the current standards, such metadata is often missing or incorrectly entered making search difficult or impossible. In this paper, we investigate automating metadata generation for SCORM-complaint LOs based on user interactions with the LO and static information about LOs and users. Our framework, called the Intelligent Learning Object Guide (iLOG), involves real-time tracking of each user sessions (an LO Wrapper), offline data mining to identify key attributes or patterns on how the LOs have been used as well as characteristics of the users (MetaGen), and the selection of these findings as metadata. Mechanisms used in the data mining include data imputation via clustering, association rule mining, and feature selection ensemble. This paper describes the methodology of automatic annotation, presents the results on the evaluation and validation of the algorithms, and discusses the resulting metadata. We have deployed our iLOG implementation for five LOs in introductory computer science topics and collected data for over 1400 sessions. We demonstrate that iLOG successfully tracks user interactions that can be used to automate the generation of meaningful empirical usage metadata for different stakeholder groups including learners and instructors, LO developers, and researchers.

KW - Association rule mining

KW - Data imputation

KW - Empirical usage metadata

KW - Feature selection

KW - SCORM learning objects

UR - http://www.scopus.com/inward/record.url?scp=84863325429&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863325429&partnerID=8YFLogxK

U2 - 10.3233/JAI-2012-024

DO - 10.3233/JAI-2012-024

M3 - Article

VL - 21

SP - 215

EP - 236

JO - International Journal of Artificial Intelligence in Education

JF - International Journal of Artificial Intelligence in Education

SN - 1560-4292

IS - 3

ER -