Title :
A Modified Constrained Random Selection Approach for Relevance Feedback
Author :
Zhang Haiyang ; Li Xianwei
Author_Institution :
Dept. of Geogr. & Environ., Suzhou Univ., Suzhou, China
Abstract :
A modified constrained random selection approach for relevance feedback is proposed in this paper. The modified method improves the representative of the negative samples, and solves the "small size sample problem" at certain degrees. Experiments demonstrated that better feedback results have got within less feedback times, comparing with original constrained random selection approach.
Keywords :
constraint handling; content-based retrieval; image retrieval; relevance feedback; modified constrained random selection approach; negative samples; relevance feedback; small size sample problem; Computer networks; Educational institutions; Feedback; Geography; Image retrieval; Information retrieval; Radio frequency; Stability; Support vector machine classification; Support vector machines;
Conference_Titel :
Photonics and Optoelectronic (SOPO), 2010 Symposium on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-4963-7
Electronic_ISBN :
978-1-4244-4964-4
DOI :
10.1109/SOPO.2010.5504356
