| | |
Summary: Tag-Based Social Image Retrieval: An Empirical Evaluation
Aixin Sun, Sourav S. Bhowmick, Khanh Tran Nam Nguyen
School of Computer Engineering, Nanyang Technological University, Singapore 639798
axsun|assourav@ntu.edu.sg
Ge Bai
School of Computer Science, Fudan University, Shanghai, China 200433
Abstract
Tags associated with social images are valuable information
source for superior image search and retrieval experiences. Al-
though various heuristics are valuable to boost tag-based search
for images, there is a lack of general framework to study the
impact of these heuristics. Specifically, the task of ranking im-
ages matching a given tag query based on their associated tags
in descending order of relevance has not been well studied. In
this paper, we take the first step to propose a generic, flexi-
ble, and extensible framework for this task and exploit it for a
systematic and comprehensive empirical evaluation of various
methods for ranking images. To this end, we identified five or-
thogonal dimensions to quantify the matching score between a
tagged image and a tag query. These five dimensions are: (i) tag
|
