i-TagRanker: an efficient tag ranking system for image sharing and retrieval using the semantic relationships between tags
- Authors
- Jeong, Jin-Woo; Hong, Hyun-Ki; Lee, Dong-Ho
- Issue Date
- Jan-2013
- Publisher
- SPRINGER
- Keywords
- Tag ranking; Tag-based image retrieval; Semantic relationship; Folksonomy; WordNet
- Citation
- MULTIMEDIA TOOLS AND APPLICATIONS, v.62, no.2, pp.451 - 478
- Indexed
- SCIE
SCOPUS
- Journal Title
- MULTIMEDIA TOOLS AND APPLICATIONS
- Volume
- 62
- Number
- 2
- Start Page
- 451
- End Page
- 478
- URI
- https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/29238
- DOI
- 10.1007/s11042-011-0903-1
- ISSN
- 1380-7501
- Abstract
- Folksonomy, considered a core component for Web 2.0 user-participation architecture, is a classification system made by user's tags on the web resources. Recently, various approaches for image retrieval exploiting folksonomy have been proposed to improve the result of image search. However, the characteristics of the tags such as semantic ambiguity and non-controlledness limit the effectiveness of tags on image retrieval. Especially, tags associated with images in a random order do not provide any information about the relevance between a tag and an image. In this paper, we propose a novel image tag ranking system called i-TagRanker which exploits the semantic relationships between tags for re-ordering the tags according to the relevance with an image. The proposed system consists of two phases: 1) tag propagation phase, 2) tag ranking phase. In tag propagation phase, we first collect the most relevant tags from similar images, and then propagate them to an untagged image. In tag ranking phase, tags are ranked according to their semantic relevance to the image. From the experimental results on a Flickr photo collection about over 30,000 images, we show the effectiveness of the proposed system.
- Files in This Item
-
Go to Link
- Appears in
Collections - COLLEGE OF COMPUTING > DEPARTMENT OF ARTIFICIAL INTELLIGENCE > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/erica/handle/2021.sw.erica/29238)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.