Comparing and evaluating information retrieval algorithms for news recommendation

Toine Bogers; Antal Van Den Bosch

doi:10.1145/1297231.1297256

Comparing and evaluating information retrieval algorithms for news recommendation

Toine Bogers, Antal Van Den Bosch

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

In this paper, we argue that the performance of content-based news recommender systems has been hampered by using relatively old and simple matching algorithms. Using more current probabilistic retrieval algorithms results in significant performance boosts. We test our ideas on a test collection that we have made publicly available. We perform both binary and graded evaluation of our algorithms and argue for the need for more graded evaluation of content-based recommender systems.

Original language	English
Title of host publication	RecSys'07 : Proceedings of the 2007 ACM Conference on Recommender Systems
Number of pages	4
Publication date	1 Dec 2007
Pages	141-144
ISBN (Print)	9781595937308
DOIs	https://doi.org/10.1145/1297231.1297256
Publication status	Published - 1 Dec 2007
Externally published	Yes
Event	RecSys'07: 2007 1st ACM Conference on Recommender Systems - Minneapolis, MN, United States Duration: 19 Oct 2007 → 20 Oct 2007

Conference

Conference	RecSys'07: 2007 1st ACM Conference on Recommender Systems
Country/Territory	United States
City	Minneapolis, MN
Period	19/10/2007 → 20/10/2007
Sponsor

Keywords

Content-based recommender systems
Probabilistic retrieval algorithms
Performance evaluation
Graded evaluation
Information retrieval

Access to Document

10.1145/1297231.1297256

Cite this

@inproceedings{8726a2be8d3f4c97bdc871ba17793b5e,

title = "Comparing and evaluating information retrieval algorithms for news recommendation",

abstract = "In this paper, we argue that the performance of content-based news recommender systems has been hampered by using relatively old and simple matching algorithms. Using more current probabilistic retrieval algorithms results in significant performance boosts. We test our ideas on a test collection that we have made publicly available. We perform both binary and graded evaluation of our algorithms and argue for the need for more graded evaluation of content-based recommender systems.",

keywords = "Content-based recommender systems, Probabilistic retrieval algorithms, Performance evaluation, Graded evaluation, Information retrieval, Content-based recommender systems, Probabilistic retrieval algorithms, Performance evaluation, Graded evaluation, Information retrieval",

author = "Toine Bogers and {Van Den Bosch}, Antal",

year = "2007",

month = dec,

day = "1",

doi = "10.1145/1297231.1297256",

language = "English",

isbn = "9781595937308",

pages = "141--144",

booktitle = "RecSys'07",

note = "RecSys'07: 2007 1st ACM Conference on Recommender Systems ; Conference date: 19-10-2007 Through 20-10-2007",

}

TY - GEN

T1 - Comparing and evaluating information retrieval algorithms for news recommendation

AU - Bogers, Toine

AU - Van Den Bosch, Antal

PY - 2007/12/1

Y1 - 2007/12/1

N2 - In this paper, we argue that the performance of content-based news recommender systems has been hampered by using relatively old and simple matching algorithms. Using more current probabilistic retrieval algorithms results in significant performance boosts. We test our ideas on a test collection that we have made publicly available. We perform both binary and graded evaluation of our algorithms and argue for the need for more graded evaluation of content-based recommender systems.

AB - In this paper, we argue that the performance of content-based news recommender systems has been hampered by using relatively old and simple matching algorithms. Using more current probabilistic retrieval algorithms results in significant performance boosts. We test our ideas on a test collection that we have made publicly available. We perform both binary and graded evaluation of our algorithms and argue for the need for more graded evaluation of content-based recommender systems.

KW - Content-based recommender systems

KW - Probabilistic retrieval algorithms

KW - Performance evaluation

KW - Graded evaluation

KW - Information retrieval

KW - Content-based recommender systems

KW - Probabilistic retrieval algorithms

KW - Performance evaluation

KW - Graded evaluation

KW - Information retrieval

UR - http://www.scopus.com/inward/record.url?scp=42149195033&partnerID=8YFLogxK

U2 - 10.1145/1297231.1297256

DO - 10.1145/1297231.1297256

M3 - Article in proceedings

AN - SCOPUS:42149195033

SN - 9781595937308

SP - 141

EP - 144

BT - RecSys'07

T2 - RecSys'07: 2007 1st ACM Conference on Recommender Systems

Y2 - 19 October 2007 through 20 October 2007

ER -

Comparing and evaluating information retrieval algorithms for news recommendation

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this