IBM / AT&T / LIG / M Lc / 3M / Submitted Systems / /
Country
Germany / United States / Singapore / /
Facility
Columbia University / University of WisconsinMadison / /
IndustryTerm
thin web client / Web Domain models / web training data / grid search / search sifier / Web Best Overall Run Run / text search / search box / text search results / meta-level learning algorithms / interactive search submissions / retrieved web images / neighbor search / semi-supervised learning algorithm / web images / web concepts / web user / Search tasks / interactive search runs / interactive search / search algorithms / nearest-neighbor search / search strateEvaluation / Web Concepts x1 x2 / semantic search engine / web search engines / search functionalities / web-train classifiers / web resources / web models / web domain / web data / grid search strategy / search results / interactive search results / learning algorithms / /
MarketIndex
set 200 / /
NaturalFeature
ASR/MT / /
OperatingSystem
L3 / /
Organization
Columbia University / US Government / U.S. Securities and Exchange Commission / University of WisconsinMadison / Team / /
Person
Michele Merler / Rong Yan / Jun Yang / Jelena Te / Wei Jiang / J. R. Smith / Semi-Supervised Learning / Vector Machines / Lexing Xie / John R. Smith / /
IBM Research TRECVID-2008 Video Retrieval System Apostol Natsev∗, Wei Jiang†, Michele Merler†, John R. Smith∗, Jelena Teˇsi´c∗, Lexing Xie∗, Rong Yan∗ Abstract In this paper, we describe the IBM Research