Automated document processing system
AssigneeHewlett-Packard Development Company, L.P.
InventorsDaniel Ortega, Sherif Yacoub, Jose Abad Peiro, Paolo Faraboschi
Life of this patent
Estimating the term…
Abstract
An automated document processing system is configured to normalize zones obtained from a document, and to extract articles from the normalized zones. In one configuration, the system receives at least one zone from the document, and applies at least one zone-breaking factor, thereby creating normalized sub-zones within which text lines are consistent with the at least one zone-breaking factor. The normalized sub-zones may be evaluated to obtain a reading order. Adjacent sub-zones are joined if text similarity exceeds a threshold value. Weakly joined sub-zones are separated where indicated by a topic vectors analysis of the weakly joined sub-zones.
Claims7
One or more computer-readable non-transitory storage media comprising computer-executable instructions for configuring a computer to extract an article from a document, the computer-executable instructions comprising instructions for:
- evaluating zones within a page of a document to obtain a reading order;
- joining adjacent zones within the reading order, where appropriate, in view of text similarity;
- and
- breaking weakly joined zones using topic vectors analysis.
A method, performed by a computer, to extract an article from a document, the method comprising:
- evaluating zones within a page of a document to obtain a reading order;
- joining adjacent zones within the reading order, where appropriate, in view of text similarity;
- and
- breaking weakly joined zones using topic vectors analysis.
A computer to extract an article from a document, the computer comprising an article extraction tool comprising:
- a zone sequencer for evaluating zones within a page of a document to obtain a reading order;
- a text correlation analyzer for joining adjacent zones within the reading order, where appropriate, in view of text similarity;
- and
- a topic vector analyzer for breaking weakly joined zones using topic vectors analysis based on a topic vector database.
Dependencies are inferred from the claim text.