4 Hits in 9.0 sec

UNED@ImageCLEF 2004: Using Image Captions Structure and Noun Phrase Based Query Expansion for Cross-Language Image Caption Retrieval

Víctor Peinado, Javier Artiles, Fernando López-Ostenero, Julio Gonzalo, Felisa Verdejo
2004 Conference and Labs of the Evaluation Forum  
Two different strategies are attempted: a) Expanding and translating queries with noun phrases, and b) Performing structured searches using entities located in topic titles over image caption fields.  ...  On the other hand, noun phrases expansion, which had proved successful in interactive tasks, influences negatively the results, due to the mismatch between the training corpus and the ImageCLEF collection  ...  Acknowledgements This work has been partially supported by a grant from the Spanish Government, project R2D2 (TIC2003-07158-C0401), and a grant from the UNED (Universidad Nacional de Educación a Distancia  ... 
dblp:conf/clef/PeinadoALGV04 fatcat:vhapfnorgvfnfbeukhkllpzvae

Semantic Mapping in Video Retrieval

Maaike H.T. de Boer
2018 SIGIR Forum  
Common knowledge bases do not always perform worse than expert knowledge, which provides options for automatic query expansion from the Web in complex event detection.  ...  We expect that query expansion has a positive effect on performance, especially if the main noun of the query cannot be detected with the available concept detectors.  ... 
doi:10.1145/3190580.3190606 fatcat:a7agjytxhng4na47sfv7xsoy2a

Dagstuhl Reports, Volume 7, Issue 7, July 2017, Complete Issue [article]

We develop a tool named Cyber-Detect, which is based on the reconstruction of an high level semantics for the binary code.  ...  Observed: In many analytic scenarios the summaries and stories presented to the user contain well-defined named entities, i.e. topics, people, places and organizations.  ...  Methods for Topicality Detection Extract the salient terms (keyterms) and entities of the document. Compare those terms to the terms found in recent news or publications, or search engine queries.  ... 
doi:10.4230/dagrep.7.7 fatcat:ve4n3wvvk5bkfgae36nsto7sre

Learning from Multimodal Web Data

John Miles Hessel
controlling for important, non-content factors like post timing.  ...  Learning from large, unlabeled web corpora has proven effective for a variety of multimodal understanding tasks.  ...  But bridging the gap between content recognition (e.g., object detection in images or named-entity recognition in text) and contextual understanding remains a challenge.  ... 
doi:10.7298/fzce-qv86 fatcat:limoc6b6xjgm5b2dbzh3f72tuq