Journal
ONLINE INFORMATION REVIEW
Volume 37, Issue 4, Pages 581-601Publisher
EMERALD GROUP PUBLISHING LTD
DOI: 10.1108/OIR-10-2011-0172
Keywords
Information retrieval; Digital libraries; Programming and algorithm theory; Recommender systems; Citation networks; Coauthorship networks
Ask authors/readers for more resources
Purpose - The scientific literature has played an important role in the dissemination of new knowledge throughout the past century. However, the increasing numbers of scientific articles being published in recent years has intensified the perception of information overload for users attempting to find relevant scientific information. The purpose of this paper is to describe a task-focused strategy that employs the task profiles of users to make recommendations in a digital library. Design/method/approach - This paper combines information retrieval, common citation analysis, and coauthor relationship analysis techniques with a citation network analysis technique the CiteRank algorithm to find relevant and high-quality articles. In total, nine variations of the proposed approach were tested using articles downloaded from the CiteSeerX system and usage logs collected from the authors' experimental server. Findings - The results from the authors' experimental evaluations demonstrate that the proposed Content-citation approach outperforms the Relevance-CiteRank, Relevance-citation count, and Relevance-only approaches. Originality/value - This paper describes an original study that has produced a novel way to combine information retrieval, common citation analysis, and coauthor relationship analysis techniques to find relevant and high-quality articles for recommendation in a digital library.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available