Classifier indexing
AssigneeHewlett Packard Enterprise Development LP
InventorsGeorge Forman, Shyamsundar Rajaram
Life of this patent
Estimating the term…
Abstract
Provided are, among other things, systems, methods and techniques for document-based processing. In one implementation, a document is input; features are extracted from it; an index is queried using at least a subset of the extracted features and, in response, identifications for selected document classifiers are received from a larger pool of document classifiers; the document is processed using individual ones of the selected document classifiers, thereby generating corresponding classifier outputs; and then, based on such classifier outputs, (1) the document is categorized within a computer database and/or (2) feedback information is provided to a user.
Claims24
A method of document-based processing, comprising:
- inputting a document;
- extracting features from the document;
- querying an index using at least a subset of the extracted features and, in response, receiving identifications for selected document classifiers from a larger pool of document classifiers;
- processing, by a system having a central processing unit, the document using individual ones of the selected document classifiers, thereby generating corresponding classifier outputs;
- and
- based on the classifier outputs, at least one of (1) categorizing the document and (2) providing feedback information to a user.
A method of information-based processing, comprising:
- receiving a set of features;
- querying an index using at least a subset of the features, and selecting, based on at least the subset of the features, classifiers from a larger pool of supervised classifiers identified by the index;
- processing, by a system having a central processing unit, the set of features by individual ones of the selected classifiers, and generating corresponding classifier outputs in response to the processing;
- and
- based on the classifier outputs, selecting at least one category from among a plurality of categories.
A non-transitory computer-readable medium storing computer-executable instructions that upon execution cause a system having a processor to:
- extract features from a document;
- query an index using at least a subset of the extracted features, and select, based on at least the subset of the extracted features, document classifiers from a larger pool of document classifiers identified by the index;
- process the document using individual ones of the selected document classifiers, and generate corresponding classifier outputs in response to the processing;
- and
- based on the classifier outputs, at least one of (1) categorize the document and (2) provide feedback information to a user.
Dependencies are inferred from the claim text.