466 search results - page 35 / 94 » Scalable Feature Extraction from Noisy Documents |
COLING
2010
13 years 4 months ago
2010
Fact collections are mostly built using semi-supervised relation extraction techniques and wisdom of the crowds methods, rendering them inherently noisy. In this paper, we propose...
ICIP
2004
IEEE
14 years 11 months ago
2004
IEEE
An object seen from different viewpoints results in differently deformed images. Affine-invariant shape classification must classify correctly the object, disregarding its viewpoi...
COLING
2002
13 years 9 months ago
2002
Automatic text categorization is a problem of automatically assigning text documents to predefined categories. In order to classify text documents, we must extract good features f...
PAKDD
2009
ACM
14 years 4 months ago
2009
ACM
Abstract. Newistic is a web mining platform that collects and analyses documents crawled from the Internet. Although it currently processes news articles, it can be easily adapted ...
KDD
2004
ACM
14 years 10 months ago
2004
ACM
Abstract. Current text classification systems typically use term stems for representing document content. Semantic Web technologies allow the usage of features on a higher semantic...