IP Library › Granted Patent US 8,442,964
Granted Patent B2
US 8,442,964 · App. 12/982,315 · Granted May 14, 2013

Information retrieval based on partial machine recognition of the same

Inventors: Rami B. Safadi (Vienna, VA); Ibrahim M. Hallaj (Waltham, MA)
Assignee: Rami B. Safadi
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 8,442,964
App. No.
12/982,315
Granted
May 14, 2013
Kind
B2
Abstract

A system and method for capturing and recognizing at least a portion of a source document, whether written or audible, then searching for information, or other documents, that correspond to the captured and recognized portion of the source document. Various techniques for adding translation and/or searching are also disclosed. In some instances, an iterative machine learning process is applied to improve the performance of an aspect of the system.

Claims (25)

1. A method for information processing in a processing system, comprising:

receiving a portion of a first complete document in a first language, said portion including some but not all information in said first complete document;

searching a database for and retrieving a complete copy of said first complete document in said first language;

machine translating said first complete document from said first language to a second language thereby generating a machine-translated document in said second language; and

using the machine-translated document in said second language as a basis for searching for a complete original document in said second language, corresponding to said machine-translated document.

2. The method of claim 1 , further comprising optical character recognition (OCR) of said portion of said first complete document and using a product of said OCR to search said database.

3. The method of claim 1 , further comprising automatic speech recognition (ASR) of said portion of said first complete document and using a product of said ASR to search said database.

4. The method of claim 1 , further comprising retrieving said complete original document in said second language, and using said complete original document to train said processing system and improve a search algorithm programmed therein.

5. The method of claim 1 , further comprising retrieving said complete original document in said second language, and using said complete original document to train said processing system and improve a machine translation algorithm programmed therein.

6. The method of claim 1 , receiving said portion comprising receiving an incomplete portion of the first complete document.

7. The method of claim 1 , receiving said portion comprising receiving an electronic text portion of the first complete document.

8. The method of claim 1 , receiving said portion comprising receiving an image representation of a portion of the first complete document.

9. The method of claim 1 , receiving said portion comprising receiving an audio representation of a portion of the first complete document.

10. The method of claim 1 , further comprising selectively choosing portions of said machine-translated document to perform the search for the complete original document.

11. The method of claim 10 , selectively choosing comprising taking selected words from said machine-translated document to perform the search for the complete original document.

12. A computer-based information processing system, comprising:

a processor for processing information including instructions storable in a memory;

said memory including machine readable instructions executable in said processor for receiving an incomplete portion of a first document in a first language;

said memory including machine readable instructions executable in said processor for searching a database for and retrieving a complete copy of said first document in said first language based on said incomplete portion;

said memory including machine readable instructions executable in said processor for translating said first document from said first language to a second language thereby generating a machine-translated document in said second language; and

said memory including machine readable instructions executable in said processor for using the machine-translated document in said second language as a basis for searching for a complete original document in said second language, corresponding to said machine-translated document.

13. A computer-implemented search method, comprising:

performing a first search using an incomplete portion of a document in a first language to locate a corresponding complete document in said first language;

performing a machine translation of said complete document in said first language to generate a machine-translated document in a second language; and

performing a second search using said machine-translated document in said second language to locate a corresponding complete original document in said second language.

Assignments (4)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 20, 2025
From: CADENA PLATFORMS LLC
To: IDENTITY DIGITAL INC.
Reel/Frame 072973/0252 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 3, 2024
From: CADENA PLATFORMS, INC.
To: CADENA PLATFORMS LLC
Reel/Frame 066008/0693 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 2, 2023
From: SAFADI, RAMI B., DR.
To: CADENA PLATFORMS, INC.
Reel/Frame 064463/0543 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 6, 2013
From: HALLAJ, IBRAHIM M
To: SAFADI, RAMI B, DR.
Reel/Frame 029761/0804 →
Continuity (2)
Provisional Application 61284995 · Dec 30, 2009
Related Publication 20110161305A1 · Jun 30, 2011