Personalized, interactive tag recommendation for flickr

Nikhil Garg, Ingmar Weber

Research output: Chapter in Book/Report/Conference proceedingConference contribution

123 Citations (Scopus)

Abstract

We study the problem of personalized, interactive tag recommendation for Flickr: While a user enters/selects new tags for a particular picture, the system suggests related tags to her, based on the tags that she or other people have used in the past along with (some of) the tags already entered. The suggested tags are dynamically updated with every additional tag entered/selected. We describe a new algorithm, called Hybrid, which can be applied to this problem, and show that it outperforms previous algorithms. It has only a single tunable parameter, which we found to be very robust. Apart from this new algorithm and its detailed analysis, our main contributions are (i) a clean methodology which leads to conservative performance estimates, (ii) showing how classical classification algorithms can be applied to this problem, (iii) introducing a new cost measure, which captures the effort of the whole tagging process, (iv) clearly identifying, when purely local schemes (using only a user's tagging history) can or cannot be improved by global schemes (using everybody's tagging history).

Original languageEnglish
Title of host publicationRecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems
Pages67-74
Number of pages8
DOIs
Publication statusPublished - 1 Dec 2008
Externally publishedYes
Event2008 2nd ACM International Conference on Recommender Systems, RecSys'08 - Lausanne, Switzerland
Duration: 23 Oct 200825 Oct 2008

Other

Other2008 2nd ACM International Conference on Recommender Systems, RecSys'08
CountrySwitzerland
CityLausanne
Period23/10/0825/10/08

Fingerprint

Costs

Keywords

  • Flickr
  • Tag co-occurrence
  • Tag recommendation
  • Tagging systems

ASJC Scopus subject areas

  • Information Systems
  • Control and Systems Engineering

Cite this

Garg, N., & Weber, I. (2008). Personalized, interactive tag recommendation for flickr. In RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems (pp. 67-74) https://doi.org/10.1145/1454008.1454020

Personalized, interactive tag recommendation for flickr. / Garg, Nikhil; Weber, Ingmar.

RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems. 2008. p. 67-74.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Garg, N & Weber, I 2008, Personalized, interactive tag recommendation for flickr. in RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems. pp. 67-74, 2008 2nd ACM International Conference on Recommender Systems, RecSys'08, Lausanne, Switzerland, 23/10/08. https://doi.org/10.1145/1454008.1454020
Garg N, Weber I. Personalized, interactive tag recommendation for flickr. In RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems. 2008. p. 67-74 https://doi.org/10.1145/1454008.1454020
Garg, Nikhil ; Weber, Ingmar. / Personalized, interactive tag recommendation for flickr. RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems. 2008. pp. 67-74
@inproceedings{470b92382d9344e9bef876535e8cbd2c,
title = "Personalized, interactive tag recommendation for flickr",
abstract = "We study the problem of personalized, interactive tag recommendation for Flickr: While a user enters/selects new tags for a particular picture, the system suggests related tags to her, based on the tags that she or other people have used in the past along with (some of) the tags already entered. The suggested tags are dynamically updated with every additional tag entered/selected. We describe a new algorithm, called Hybrid, which can be applied to this problem, and show that it outperforms previous algorithms. It has only a single tunable parameter, which we found to be very robust. Apart from this new algorithm and its detailed analysis, our main contributions are (i) a clean methodology which leads to conservative performance estimates, (ii) showing how classical classification algorithms can be applied to this problem, (iii) introducing a new cost measure, which captures the effort of the whole tagging process, (iv) clearly identifying, when purely local schemes (using only a user's tagging history) can or cannot be improved by global schemes (using everybody's tagging history).",
keywords = "Flickr, Tag co-occurrence, Tag recommendation, Tagging systems",
author = "Nikhil Garg and Ingmar Weber",
year = "2008",
month = "12",
day = "1",
doi = "10.1145/1454008.1454020",
language = "English",
isbn = "9781605580937",
pages = "67--74",
booktitle = "RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems",

}

TY - GEN

T1 - Personalized, interactive tag recommendation for flickr

AU - Garg, Nikhil

AU - Weber, Ingmar

PY - 2008/12/1

Y1 - 2008/12/1

N2 - We study the problem of personalized, interactive tag recommendation for Flickr: While a user enters/selects new tags for a particular picture, the system suggests related tags to her, based on the tags that she or other people have used in the past along with (some of) the tags already entered. The suggested tags are dynamically updated with every additional tag entered/selected. We describe a new algorithm, called Hybrid, which can be applied to this problem, and show that it outperforms previous algorithms. It has only a single tunable parameter, which we found to be very robust. Apart from this new algorithm and its detailed analysis, our main contributions are (i) a clean methodology which leads to conservative performance estimates, (ii) showing how classical classification algorithms can be applied to this problem, (iii) introducing a new cost measure, which captures the effort of the whole tagging process, (iv) clearly identifying, when purely local schemes (using only a user's tagging history) can or cannot be improved by global schemes (using everybody's tagging history).

AB - We study the problem of personalized, interactive tag recommendation for Flickr: While a user enters/selects new tags for a particular picture, the system suggests related tags to her, based on the tags that she or other people have used in the past along with (some of) the tags already entered. The suggested tags are dynamically updated with every additional tag entered/selected. We describe a new algorithm, called Hybrid, which can be applied to this problem, and show that it outperforms previous algorithms. It has only a single tunable parameter, which we found to be very robust. Apart from this new algorithm and its detailed analysis, our main contributions are (i) a clean methodology which leads to conservative performance estimates, (ii) showing how classical classification algorithms can be applied to this problem, (iii) introducing a new cost measure, which captures the effort of the whole tagging process, (iv) clearly identifying, when purely local schemes (using only a user's tagging history) can or cannot be improved by global schemes (using everybody's tagging history).

KW - Flickr

KW - Tag co-occurrence

KW - Tag recommendation

KW - Tagging systems

UR - http://www.scopus.com/inward/record.url?scp=63449122298&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=63449122298&partnerID=8YFLogxK

U2 - 10.1145/1454008.1454020

DO - 10.1145/1454008.1454020

M3 - Conference contribution

AN - SCOPUS:63449122298

SN - 9781605580937

SP - 67

EP - 74

BT - RecSys'08: Proceedings of the 2008 ACM Conference on Recommender Systems

ER -