Abstract

Automatically recommending suitable tags for online content is a necessary task for better information organization and retrieval. In this article, we propose a generative model SIMWORD for the tag recommendation problem on textual content. The key observation of our model is that the tags and their relevant/similar words may have appeared in the corresponding content. In particular, we first empirically verify this observation in real data sets, and then design a supervised topic model which is guided by the above observation for tag recommendation. Experimental evaluations demonstrate that the proposed method outperforms several existing methods in terms of recommendation accuracy.

Original languageEnglish (US)
Pages (from-to)479-489
Number of pages11
JournalNeurocomputing
Volume314
DOIs
StatePublished - Nov 7 2018

Fingerprint

Observation
Information Storage and Retrieval
Datasets

Keywords

  • Generative model
  • Relevant words
  • Similar words
  • Supervised topic modeling
  • Tag recommendation

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

Guiding supervised topic modeling for content based tag recommendation. / Wu, Yong; Xi, Shengqu; Yao, Yuan; Xu, Feng; Tong, Hanghang; Lu, Jian.

In: Neurocomputing, Vol. 314, 07.11.2018, p. 479-489.

Research output: Contribution to journalArticle

Wu, Yong ; Xi, Shengqu ; Yao, Yuan ; Xu, Feng ; Tong, Hanghang ; Lu, Jian. / Guiding supervised topic modeling for content based tag recommendation. In: Neurocomputing. 2018 ; Vol. 314. pp. 479-489.
@article{b748e232d5c445d1933f5539adbe92cf,
title = "Guiding supervised topic modeling for content based tag recommendation",
abstract = "Automatically recommending suitable tags for online content is a necessary task for better information organization and retrieval. In this article, we propose a generative model SIMWORD for the tag recommendation problem on textual content. The key observation of our model is that the tags and their relevant/similar words may have appeared in the corresponding content. In particular, we first empirically verify this observation in real data sets, and then design a supervised topic model which is guided by the above observation for tag recommendation. Experimental evaluations demonstrate that the proposed method outperforms several existing methods in terms of recommendation accuracy.",
keywords = "Generative model, Relevant words, Similar words, Supervised topic modeling, Tag recommendation",
author = "Yong Wu and Shengqu Xi and Yuan Yao and Feng Xu and Hanghang Tong and Jian Lu",
year = "2018",
month = "11",
day = "7",
doi = "10.1016/j.neucom.2018.07.011",
language = "English (US)",
volume = "314",
pages = "479--489",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Guiding supervised topic modeling for content based tag recommendation

AU - Wu, Yong

AU - Xi, Shengqu

AU - Yao, Yuan

AU - Xu, Feng

AU - Tong, Hanghang

AU - Lu, Jian

PY - 2018/11/7

Y1 - 2018/11/7

N2 - Automatically recommending suitable tags for online content is a necessary task for better information organization and retrieval. In this article, we propose a generative model SIMWORD for the tag recommendation problem on textual content. The key observation of our model is that the tags and their relevant/similar words may have appeared in the corresponding content. In particular, we first empirically verify this observation in real data sets, and then design a supervised topic model which is guided by the above observation for tag recommendation. Experimental evaluations demonstrate that the proposed method outperforms several existing methods in terms of recommendation accuracy.

AB - Automatically recommending suitable tags for online content is a necessary task for better information organization and retrieval. In this article, we propose a generative model SIMWORD for the tag recommendation problem on textual content. The key observation of our model is that the tags and their relevant/similar words may have appeared in the corresponding content. In particular, we first empirically verify this observation in real data sets, and then design a supervised topic model which is guided by the above observation for tag recommendation. Experimental evaluations demonstrate that the proposed method outperforms several existing methods in terms of recommendation accuracy.

KW - Generative model

KW - Relevant words

KW - Similar words

KW - Supervised topic modeling

KW - Tag recommendation

UR - http://www.scopus.com/inward/record.url?scp=85050380118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85050380118&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2018.07.011

DO - 10.1016/j.neucom.2018.07.011

M3 - Article

VL - 314

SP - 479

EP - 489

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -