A Subspace Clustering Framework for Research Group Collaboration

Nitin Agarwal, Ehtesham Haque, Huan Liu, Lance Parsons

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Researchers spend considerable time searching for relevant papers on the topic in which they are currently interested. Often, despite having similar interests, researchers in the same laboratory do not find it convenient to share results of bibliographic searches and thus conduct independent time-consuming searches. Research paper recommender systems can help the researcher avoid such time-consuming searches by allowing each researcher to automatically take advantage of previous searches performed by others in the lab. Existing recommender systems were developed for commercial domains to assist users by focusing toward products of their interests. Unlike those domains, the research paper domain has relatively few users when compared with the significantly larger number of research papers. In this paper, we present a novel system to recommend relevant research papers to a user based on the user’s recent querying and browsing habits. The core of the system is a scalable subspace clustering algorithm, SCuBA (Subspace ClUstering Based Analysis) that performs well on the sparse, high-dimensional data collected in this domain. Both synthetic and benchmark datasets are used to evaluate the recommendation system and to demonstrate that it performs better than the traditional collaborative filtering approaches when recommending research papers.

Original languageEnglish (US)
Pages (from-to)35-58
Number of pages24
JournalInternational Journal of Information Technology and Web Engineering
Volume1
Issue number1
DOIs
StatePublished - 2006

Fingerprint

Recommender systems
Collaborative filtering
Clustering algorithms

Keywords

  • collaborative filtering
  • high-dimensional data
  • highly sparse data
  • recommender systems
  • research paper domain
  • subspace clustering

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

A Subspace Clustering Framework for Research Group Collaboration. / Agarwal, Nitin; Haque, Ehtesham; Liu, Huan; Parsons, Lance.

In: International Journal of Information Technology and Web Engineering, Vol. 1, No. 1, 2006, p. 35-58.

Research output: Contribution to journalArticle

@article{2d09d6b04df94ef5a1094381c0a6dc8b,
title = "A Subspace Clustering Framework for Research Group Collaboration",
abstract = "Researchers spend considerable time searching for relevant papers on the topic in which they are currently interested. Often, despite having similar interests, researchers in the same laboratory do not find it convenient to share results of bibliographic searches and thus conduct independent time-consuming searches. Research paper recommender systems can help the researcher avoid such time-consuming searches by allowing each researcher to automatically take advantage of previous searches performed by others in the lab. Existing recommender systems were developed for commercial domains to assist users by focusing toward products of their interests. Unlike those domains, the research paper domain has relatively few users when compared with the significantly larger number of research papers. In this paper, we present a novel system to recommend relevant research papers to a user based on the user’s recent querying and browsing habits. The core of the system is a scalable subspace clustering algorithm, SCuBA (Subspace ClUstering Based Analysis) that performs well on the sparse, high-dimensional data collected in this domain. Both synthetic and benchmark datasets are used to evaluate the recommendation system and to demonstrate that it performs better than the traditional collaborative filtering approaches when recommending research papers.",
keywords = "collaborative filtering, high-dimensional data, highly sparse data, recommender systems, research paper domain, subspace clustering",
author = "Nitin Agarwal and Ehtesham Haque and Huan Liu and Lance Parsons",
year = "2006",
doi = "10.4018/jitwe.2006010102",
language = "English (US)",
volume = "1",
pages = "35--58",
journal = "International Journal of Information Technology and Web Engineering",
issn = "1554-1045",
publisher = "IGI Publishing",
number = "1",

}

TY - JOUR

T1 - A Subspace Clustering Framework for Research Group Collaboration

AU - Agarwal, Nitin

AU - Haque, Ehtesham

AU - Liu, Huan

AU - Parsons, Lance

PY - 2006

Y1 - 2006

N2 - Researchers spend considerable time searching for relevant papers on the topic in which they are currently interested. Often, despite having similar interests, researchers in the same laboratory do not find it convenient to share results of bibliographic searches and thus conduct independent time-consuming searches. Research paper recommender systems can help the researcher avoid such time-consuming searches by allowing each researcher to automatically take advantage of previous searches performed by others in the lab. Existing recommender systems were developed for commercial domains to assist users by focusing toward products of their interests. Unlike those domains, the research paper domain has relatively few users when compared with the significantly larger number of research papers. In this paper, we present a novel system to recommend relevant research papers to a user based on the user’s recent querying and browsing habits. The core of the system is a scalable subspace clustering algorithm, SCuBA (Subspace ClUstering Based Analysis) that performs well on the sparse, high-dimensional data collected in this domain. Both synthetic and benchmark datasets are used to evaluate the recommendation system and to demonstrate that it performs better than the traditional collaborative filtering approaches when recommending research papers.

AB - Researchers spend considerable time searching for relevant papers on the topic in which they are currently interested. Often, despite having similar interests, researchers in the same laboratory do not find it convenient to share results of bibliographic searches and thus conduct independent time-consuming searches. Research paper recommender systems can help the researcher avoid such time-consuming searches by allowing each researcher to automatically take advantage of previous searches performed by others in the lab. Existing recommender systems were developed for commercial domains to assist users by focusing toward products of their interests. Unlike those domains, the research paper domain has relatively few users when compared with the significantly larger number of research papers. In this paper, we present a novel system to recommend relevant research papers to a user based on the user’s recent querying and browsing habits. The core of the system is a scalable subspace clustering algorithm, SCuBA (Subspace ClUstering Based Analysis) that performs well on the sparse, high-dimensional data collected in this domain. Both synthetic and benchmark datasets are used to evaluate the recommendation system and to demonstrate that it performs better than the traditional collaborative filtering approaches when recommending research papers.

KW - collaborative filtering

KW - high-dimensional data

KW - highly sparse data

KW - recommender systems

KW - research paper domain

KW - subspace clustering

UR - http://www.scopus.com/inward/record.url?scp=85001785546&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85001785546&partnerID=8YFLogxK

U2 - 10.4018/jitwe.2006010102

DO - 10.4018/jitwe.2006010102

M3 - Article

AN - SCOPUS:85001785546

VL - 1

SP - 35

EP - 58

JO - International Journal of Information Technology and Web Engineering

JF - International Journal of Information Technology and Web Engineering

SN - 1554-1045

IS - 1

ER -