Seungwoo Lee

From WikiPapers
Jump to: navigation, search

Seungwoo Lee is an author.


Only those publications related to wikis are shown here.
Title Keyword(s) Published in Language DateThis property is a special property in this wiki. Abstract R C
Web image retrieval re-ranking with Wikipedia semantics Flickr
Image tags
Semantic relatedness
Web image retrieval
International Journal of Multimedia and Ubiquitous Engineering English 2012 Nowadays, to take advantage of tags is a general tendency when users need to store or retrieve images on the Web. In this article, we introduce some approaches to calculate semantic importance of tags attached to Web images, and to make re-ranking the retrieved images according to them. We have compared the results from image re-ranking with two semantic providers, Word Net and Wikipedia. With the semantic importance of image tags calculated by using Wikipedia, we found the superiority of the method in precision and recall rate as experimental results. 0 0
Web image retrieval using semantic prior tags Image tags
Semantic information
Web image retrieval
Journal of Convergence Information Technology English 2012 This research is for extraction and utilization of semantic information from the Web image tags based on Wikipedia. Generally, most photo images stored on the Web have lots of tags added with user's subjective judgments, not by the importance of them. So, in tagged Web image retrieval, they have become the cause of precision rate decrease. In this paper, we suggest a method deals with selecting prior tags when tagged images are uploaded to online Web image databases, and using them in image retrieval. This method includes calculation of semantic relatedness between tags based on Wikipedia for prior tag selection. Also, it is characterized by multilevel search of tagged images with prior tags. For evaluation, we compared our method with Flickr's method, which is a simple matching of tags to a given query. As the results, we found the superiority of our method in precision and recall rate. 0 0
Measuring Similarities between Technical Terms Based on Wikipedia Similarity Measure
Technical Terms
Wikipedia InterLink
Wikipedia Category
ITHINGSCPSCOM English 2011 0 0
Measuring similarities between technical terms based on Wikipedia Similarity measure
Technical terms
Wikipedia category
Wikipedia internal link
Proceedings - 2011 IEEE International Conferences on Internet of Things and Cyber, Physical and Social Computing, iThings/CPSCom 2011 English 2011 Measuring similarities between terms is useful for semantic information processing such as query expansion and WSD (Word Sense Disambiguation). This study aims at identifying technologies closely related to emerging technologies. Thus, we propose a hybrid method using both category and internal link information in Wikipedia, which is the largest database that everyone can share and edit its contents. Comparative experimental results with a state-of-theart WLM (Wikipedia Link-based Measure) show that this proposed method works better than each single method. 0 0