The two most important tasks in entity information summarization from the Web are named entity recognition and relation extraction. Little work has been done toward an integrated ...
Human-nameable visual “attributes” can benefit various recognition tasks. However, existing techniques restrict these properties to categorical labels (for example, a person ...
Abstract. Previous works in change detection on XML focused on detecting changes to text file using ordered and unordered tree model. These approaches are not suitable for detecti...
Erwin Leonardi, Sourav S. Bhowmick, T. S. Dharma, ...
Blocking is a technique commonly used in manual statistical analysis to account for confounding variables. However, blocking is not currently used in automated learning algorithms...
Operations on spatial objects have much individuality. As a consequence, the spatial data modelling approaches, which have been proposed, have to consider either data types of the...