Relevance Feedback for Content-Based Information Retrieval
Relevance feedback (RF) [2, 5] is an on-line approach which tries to learn the user’s intentions on the fly. It leverages users to guide the computers to search for relevant documents. An RF mechanism has two components: a learner and a selector. At every feedback round, the user marks (part of) the images returned by the search engine as relevant or irrelevant. The learner exploits this information to re-estimate the target of the user. This information is used both quantitatively (retrieving more documents like the relevant documents) and qualitatively (retrieving documents similar to the relevant ones before other documents). With the current estimation of the target, the selector chooses other images that are displayed by the interface of the search engine; then the user is asked to provide feedback on these images during the next round. The process of RF is usually presented as a cycle of activity: an IR system presents a user with a set of retrieved documents; the...
- 2.Crucianu M, Ferecatu M, Boujemaa N. Relevance feedback for image retrieval: a short survey. In: State of the art in audiovisual content-based retrieval, Information Universal Access and Interaction, Including Datamodels and Languages. Report of the DELOS2 European Network of Excellence (FP6). (2004).Google Scholar
- 7.Tian Q, Yu Y, Huang TS. Incorporate discriminant analysis with EM algorithm in image retrieval. In: Proceedings of the IEEE International Conference on Multimedia and Expo; 2000. p. 299–302.Google Scholar
- 8.Tong S, Chang E. Support vector machine active learning for image retrieval. In: Proceedings of the 9th ACM International Conference on Multimedia; 2001. p. 107–18.Google Scholar