System and method for automatic document management

• US 10,204,143 B1
• Filed: 03/30/2015
• Issued: 02/12/2019
• Est. Priority Date: 11/02/2011
• Status: Active Grant

• Current Assignee: Autofile Incorporated
• Original Assignee: DUB Software Group Incorporated
• Inventors: Dub, Eitan, Dub, Adam O., Miro, Alfredo J.
• Primary Examiner(s): To, Baoquoc N
• Application Number:
  US14/673,207
• Time in Patent Office: 1,415 Days
• Field of Search: 707730, 707749, 715205, 382224, 382159
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/93 Document management systems

Abstract

A system for managing documents, comprising: interfaces to a user interface, proving an application programming interface, a database of document images, a remote server, configured to communicate a text representation of the document from the optical character recognition engine to the report server, and to receive from the remote server a classification of the document; and logic configured to receive commands from the user interface, and to apply the classifications received from the remote server to the document images through the interface to the database. A corresponding method is also provided.

Claims

1. A method for managing documents, comprising:

scanning a series of pages representing a plurality of documents;

performing optical character recognition on the series of pages to produce semantic content;

communicating the semantic content of the series of pages as untagged and unseparated information;

analyzing the semantic content of the series of pages to produce;

a document structure of the untagged and unseparated information as a series of respective document portions, and

a status for respective document portions based on at least a correspondence of the semantic content of the respective document portion to a plurality of statistical semantic classification features, the status being selected from the group consisting of (a) a document classification of the respective document portion, and (b) an exception to classification status of the respective document portion;

prompting to receive prompted manual classification for document portions having the exception to classification status, wherein the manual classification is applied to the document portions having the exception to classification status; and

updating the plurality of statistical semantic classification features based on the received manual classification and the semantic content of the respective document portion.

2. The method according to claim 1, further comprising storing a record in a database representing:

a series pages, associated with semantic content of the series of pages derived by automated optical character recognition of an optical scan of the series of pages, and

the status for each document portion of the series of pages selected from the group consisting of;

the classification of respective document portion,

the exception to document classification status for the respective document portion, and

an unclassified status for the series of documents, the document database further comprising an index of semantic content of the records in the database.

3. The method according to claim 2, further comprising selectively accessing the records based on the index of semantic content.

4. The method according to claim 2, further comprising selectively accessing the records based on the classification of respective document portion.

5. The method according to claim 2, wherein each respective document portion has a provenance representing a source batch of page images and a place within a sequence of the batch of page images, wherein the provenance is stored in the document database in association with the record associated with a respective document portion.

6. The method according to claim 1, further comprising:

receiving user-defined classification categories;

receiving a manual classification from a user for classifying a respective document portion based on the user-defined classification categories.

7. The method according to claim 1, further comprising automatically performing at least one task selectively dependent on content of a respective document portion.

8. The method according to claim 1, further comprising reprocessing of a set of document portions according to the updated plurality of statistical semantic classification features.

11. The system according to claim 1, further comprising a database configured to store records, each representing a series pages, associated with semantic content of the series of pages, and the status for each document portion of the series of pages selected from the group consisting of the classification of respective document portion, the exception to document classification status for the respective document portion, and an unclassified status for the series of documents, the database further comprising an index of semantic content of the records in the database.

12. The system according to claim 11, wherein the database is configured to selectively access the records based on the index of semantic content.

13. The system according to claim 11, wherein the database is configured to selectively access the records based on the classification of respective document portion.

14. The system according to claim 11, wherein each respective document portion has a provenance representing a source batch of page images and a place within a sequence of the batch of page images, wherein the provenance is stored in the database in association with the record associated with a respective document portion.

9. A system for managing documents, comprising:

a memory configured to store semantic content derived from a series of pages representing a plurality of documents;

a communication port configured to communicate the semantic content as untagged and unseparated information; and

at least one processor configured to;

analyze the semantic content of the series of pages to produce;

a document structure of the untagged and unseparated information as a series of respective document portions, and

a status for respective document portions based on at least a correspondence of the semantic content of the respective document portion to a plurality of statistical semantic classification features, selected from the group consisting of (a) a document classification of the respective document portion, and (b) an exception to classification status of the respective document portion,

generate a prompt to a user to supply a manual classification for document portions having the exception to classification status, wherein the manual classification is applied to the document portions having the exception to classification status; and

update the plurality of statistical semantic classification features based on the received manual classification and the semantic content of the respective document portion.

10. The system according to claim 9, further comprising an optical character recognition engine to produce the semantic content from a series of page images representing the plurality of documents.

15. The system according to claim 9, further comprising a user interface configured to receive user-defined classification categories and a manual classification from a user for classifying a respective document portion based on the user-defined classification categories.

16. The system according to claim 9, wherein the at least one automated processor is further configured to perform at least one task selectively dependent on content of a respective document portion.

17. The system according to claim 9, wherein the at least one processor is further configured to reprocess of a set of document portions according to the updated plurality of statistical semantic classification features.

18. A method for managing documents, comprising:

storing a semantic content of a plurality documents, which is unseparated and untagged with respect to document structure, in a database, each document encompassing a series of document portions, each respective document containing semantic content derived by automated image analysis, and a status selected from the group consisting of a document portion classification, an exception to document portion classification status, and an unclassified status for the respective document;

communicating the stored semantic content which is unseparated and untagged with respect to document structure to a remote server;

receiving from the remote server a response comprising a separated and tagged document structure of respective document portions, and a status selectively based on a correspondence of the semantic content of a respective document portion to a plurality of classes based on statistical semantic classification features, selected from the group consisting of (a) a classification of a respective document, and (b) an exception to classification status for the respective document, the respective document portion being automatically classified as belonging to at least one class if the correspondence of semantic content of the respective document portion is high for the at least one class; and

the respective document portion having an exception to document classification status if a correspondence of the semantic content of the document portion is not high for any class.

19. The method according to claim 18, further comprising:

receiving a manual classification of a respective document portion; and

updating the record associated with the respective document based on the manual classification.

20. The method according to claim 18, further comprising retrieving the document portions in the database based on a semantic content index.