Title :
ImageScour: a specialized search engine
Author :
Simina, Marin ; Edelman, Adam ; Barbu, Costin
Author_Institution :
Dept. of Comput. & Inf. Sci., Loyola Univ., New Orleans, LA, USA
Abstract :
This paper presents a specialized search engine for image retrieval called ImageScour. Compared with a traditional search engine, ImageScour takes advantage of contextual and topological information about the requested image to improve its retrieval performance. In particular, ImageScour is relevant for automatically repairing "broken" links pointing to other Websites, when the content of these Websites is updated or reorganized. Empirical results show that ImageScour outperforms general search engines (e.g., the advanced Google image search) by a significant margin.
Keywords :
Internet; data mining; image retrieval; search engines; contextual information; data mining; image retrieval; intelligent system; search engine; topological information; Computational Intelligence Society; Data mining; Image analysis; Image retrieval; Indexing; Information resources; Intelligent systems; Search engines; Topology; Web pages;
Conference_Titel :
Systems, Man and Cybernetics, 2004 IEEE International Conference on
Print_ISBN :
0-7803-8566-7
DOI :
10.1109/ICSMC.2004.1399833
