Quest for relevant tags using local interaction networks and visual content

Neela Sawant, Ritendra Datta, Jia Li, James Wang

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

19 Citations (Scopus)

Abstract

Typical tag recommendation systems for photos shared on social networks such as Flickr, use visual content analysis, collaborative filtering or personalization strategies to produce annotations. However, the dependence on manual intervention and the knowledge of sufficient personal preferences coupled with the folksonomic issues limit the scope of these strategies. In this paper, we present a fully automatic and folksonomically scalable tag recommendation model that can recommend tags for a user's photos without an explicit knowledge of the user's personal tagging preferences. The model is learned using the collective tagging behavior of other users in the user's local interaction network, which we believe approximates the user's preferences, at least partially. The tag recommendation model generates content-based annotations and then uses a Nave Bayes formulation to translate these annotations to a set of folksonomic tags selected from the tags used by the users in the local interaction network. Quantitative and qualitative comparisons with 890 Flickr networks show that this approach is highly useful for tag recommendation in the presence of insufficient information of a user's own preferences.

Original languageEnglish (US)
Title of host publicationMIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval
Pages231-240
Number of pages10
DOIs
StatePublished - May 21 2010
Event2010 ACM SIGMM International Conference on Multimedia Information Retrieval, MIR 2010 - Philadelphia, PA, United States
Duration: Mar 29 2010Mar 31 2010

Publication series

NameMIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval

Other

Other2010 ACM SIGMM International Conference on Multimedia Information Retrieval, MIR 2010
CountryUnited States
CityPhiladelphia, PA
Period3/29/103/31/10

Fingerprint

Collaborative filtering
Recommender systems

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Information Systems

Cite this

Sawant, N., Datta, R., Li, J., & Wang, J. (2010). Quest for relevant tags using local interaction networks and visual content. In MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval (pp. 231-240). (MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval). https://doi.org/10.1145/1743384.1743424
Sawant, Neela ; Datta, Ritendra ; Li, Jia ; Wang, James. / Quest for relevant tags using local interaction networks and visual content. MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval. 2010. pp. 231-240 (MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval).
@inproceedings{959830f45b2c4780b8195b54ad8bcb7b,
title = "Quest for relevant tags using local interaction networks and visual content",
abstract = "Typical tag recommendation systems for photos shared on social networks such as Flickr, use visual content analysis, collaborative filtering or personalization strategies to produce annotations. However, the dependence on manual intervention and the knowledge of sufficient personal preferences coupled with the folksonomic issues limit the scope of these strategies. In this paper, we present a fully automatic and folksonomically scalable tag recommendation model that can recommend tags for a user's photos without an explicit knowledge of the user's personal tagging preferences. The model is learned using the collective tagging behavior of other users in the user's local interaction network, which we believe approximates the user's preferences, at least partially. The tag recommendation model generates content-based annotations and then uses a Nave Bayes formulation to translate these annotations to a set of folksonomic tags selected from the tags used by the users in the local interaction network. Quantitative and qualitative comparisons with 890 Flickr networks show that this approach is highly useful for tag recommendation in the presence of insufficient information of a user's own preferences.",
author = "Neela Sawant and Ritendra Datta and Jia Li and James Wang",
year = "2010",
month = "5",
day = "21",
doi = "10.1145/1743384.1743424",
language = "English (US)",
isbn = "9781605588155",
series = "MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval",
pages = "231--240",
booktitle = "MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval",

}

Sawant, N, Datta, R, Li, J & Wang, J 2010, Quest for relevant tags using local interaction networks and visual content. in MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval. MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval, pp. 231-240, 2010 ACM SIGMM International Conference on Multimedia Information Retrieval, MIR 2010, Philadelphia, PA, United States, 3/29/10. https://doi.org/10.1145/1743384.1743424

Quest for relevant tags using local interaction networks and visual content. / Sawant, Neela; Datta, Ritendra; Li, Jia; Wang, James.

MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval. 2010. p. 231-240 (MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval).

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

TY - GEN

T1 - Quest for relevant tags using local interaction networks and visual content

AU - Sawant, Neela

AU - Datta, Ritendra

AU - Li, Jia

AU - Wang, James

PY - 2010/5/21

Y1 - 2010/5/21

N2 - Typical tag recommendation systems for photos shared on social networks such as Flickr, use visual content analysis, collaborative filtering or personalization strategies to produce annotations. However, the dependence on manual intervention and the knowledge of sufficient personal preferences coupled with the folksonomic issues limit the scope of these strategies. In this paper, we present a fully automatic and folksonomically scalable tag recommendation model that can recommend tags for a user's photos without an explicit knowledge of the user's personal tagging preferences. The model is learned using the collective tagging behavior of other users in the user's local interaction network, which we believe approximates the user's preferences, at least partially. The tag recommendation model generates content-based annotations and then uses a Nave Bayes formulation to translate these annotations to a set of folksonomic tags selected from the tags used by the users in the local interaction network. Quantitative and qualitative comparisons with 890 Flickr networks show that this approach is highly useful for tag recommendation in the presence of insufficient information of a user's own preferences.

AB - Typical tag recommendation systems for photos shared on social networks such as Flickr, use visual content analysis, collaborative filtering or personalization strategies to produce annotations. However, the dependence on manual intervention and the knowledge of sufficient personal preferences coupled with the folksonomic issues limit the scope of these strategies. In this paper, we present a fully automatic and folksonomically scalable tag recommendation model that can recommend tags for a user's photos without an explicit knowledge of the user's personal tagging preferences. The model is learned using the collective tagging behavior of other users in the user's local interaction network, which we believe approximates the user's preferences, at least partially. The tag recommendation model generates content-based annotations and then uses a Nave Bayes formulation to translate these annotations to a set of folksonomic tags selected from the tags used by the users in the local interaction network. Quantitative and qualitative comparisons with 890 Flickr networks show that this approach is highly useful for tag recommendation in the presence of insufficient information of a user's own preferences.

UR - http://www.scopus.com/inward/record.url?scp=77952338762&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952338762&partnerID=8YFLogxK

U2 - 10.1145/1743384.1743424

DO - 10.1145/1743384.1743424

M3 - Conference contribution

SN - 9781605588155

T3 - MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval

SP - 231

EP - 240

BT - MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval

ER -

Sawant N, Datta R, Li J, Wang J. Quest for relevant tags using local interaction networks and visual content. In MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval. 2010. p. 231-240. (MIR 2010 - Proceedings of the 2010 ACM SIGMM International Conference on Multimedia Information Retrieval). https://doi.org/10.1145/1743384.1743424