Abstract: A classifier learning module trains video classifiers associated with a stored set of concepts derived from textual metadata of a plurality of videos. Specifically, a first type of classifier (e.g., a content-based classifier) and a second type of classifier (e.g., a text-based classifier) are trained, the classifiers when applied to a video indicating a likelihood that the video represents one or more concepts corresponding to the classifier. The first type of classifier can be used to determine the training set for the second type of classifier. The learning process does not require any concepts to be known a priori, nor does it require a training set of videos having training labels manually applied by human experts. Scores from the first type of classifier are combined with scores from the second type of classifier to obtain video classification of enhanced accuracy.

Abstract: Reference identification and resolution identifies reference text fragments in a document and associates referenced object text fragments in the document with the identified reference text fragments. Reference profiles are abstracted from the document. Each reference profile specifies at least a reference number and an object type identifier. A reference profile is paired with an object text fragment of the document containing the reference number of the reference profile. The pairing is repeated to associate reference profiles with object text fragments. A reference text fragment of the document satisfying one of the reference profiles is associated with the object text fragment paired with the satisfied reference profile. The associating is repeated to associate reference text fragments of the document with object text fragments.

Abstract: Reference identification and resolution identifies reference text fragments in a document and associates referenced object text fragments in the document with the identified reference text fragments. Reference profiles are abstracted from the document. Each reference profile specifies at least a reference number and an object type identifier. A reference profile is paired with an object text fragment of the document containing the reference number of the reference profile. The pairing is repeated to associate reference profiles with object text fragments. A reference text fragment of the document satisfying one of the reference profiles is associated with the object text fragment paired with the satisfied reference profile. The associating is repeated to associate reference text fragments of the document with object text fragments.