Christopher Dangelo Dozier

US Patent:

7333966, Feb 19, 2008

Filed:

Jun 13, 2002

Appl. No.:

10/171170

Inventors:

Christopher C. Dozier - Eagan MN, US

Assignee:

Thomson Global Resources - Zug

International Classification:

G06F 7/00
G06F 17/30
G06F 15/18
G06E 1/00
G06E 3/00

US Classification:

706 21, 707 2, 707100, 235435, 706 20, 706 16

Abstract:

Hyperlinking or associating documents to other documents based on the names of people in the documents has become more desirable. Although there is an automated system for installing such hyperlinks into judicial opinions, the system is not generally applicable to other types of names and documents, nor well suited to determine hyperlinks for names that might refer to two or more similarly named persons. Accordingly, the inventor devised systems, methods, and software that facilitate hyperlinking names in documents, regardless of type. One exemplary system includes a descriptor module and a linking module. The descriptor module develops descriptive patterns for selecting co-occurent document information that is useful in recognizing associations between names and professional classes. The linking module tags names in an input document, extracts co-occurent information using the descriptive patterns, and uses a Bayesian inference network that processes a (non-inverse-document-frequency) name-rarity score for each name along with the name and selected co-occurent document information to determine appropriate hyperlinks to other documents, such as entries in professional directories.

US Patent:

7571174, Aug 4, 2009

Filed:

Jan 3, 2005

Appl. No.:

11/028464

Inventors:

Yohendran Arumainayagam - Stamford CT, US
Christopher C. Dozier - Minneapolis MN, US

Assignee:

Thomson Reuters Global Resurces - Zug

International Classification:

G06F 17/30

US Classification:

707101, 707 4, 707 7, 707 10

Abstract:

An information-retrieval system includes a server that receives queries for documents from client devices and means for outputting results of queries to the client devices, with the results provided in association with one or more interactive control features that are selectable to invoke display of information regarding entities, such as professionals, referenced in the results.

US Patent:

8001129, Aug 16, 2011

Filed:

Jul 28, 2009

Appl. No.:

12/510914

Inventors:

Yohendran Arumainayagam - Lakeville MN, US
Christopher C. Dozier - Minneapolis MN, US

Assignee:

Thomson Reuters Global Resources - Baar

International Classification:

G06F 17/30

US Classification:

707749, 707752, 707802, 707803

Abstract:

An information-retrieval system includes a server that receives queries for documents from client devices and means for outputting results of queries to the client devices, with the results provided in association with one or more interactive control features that are selectable to invoke display of information regarding entities, such as professionals, referenced in the results.

US Patent:

20070156674, Jul 5, 2007

Filed:

Oct 4, 2006

Appl. No.:

11/538583

Inventors:

Christopher Dozier - Minneapolis MN, US
Mark Chaudhary - Eagan MN, US
Ravi Kondadadi - Eagan MN, US

Assignee:

West Services, Inc. - Eagan MN

International Classification:

G06F 17/30

US Classification:

707005000

Abstract:

Some known medical terms may function as non-medical terms depending on their particular context. Accordingly, the present inventors devised systems, methods, and software that facilitate determining whether a term that is found in a medical corpus is likely to be a medical term when found in another corpus. An exemplary embodiment receives a term and computes an ambiguity score based on language models for a medical and a non-medical corpus.

US Patent:

20080301074, Dec 4, 2008

Filed:

Dec 19, 2007

Appl. No.:

11/960054

Inventors:

Christopher C. Dozier - Eagan MN, US

International Classification:

G06E 1/00
G06F 17/00
G06F 7/00
G06F 17/30
G06E 3/00

US Classification:

706 21, 715205, 707100, 707 5, 707E17013, 707E17074

Abstract:

Hyperlinking or associating documents to other documents based on the names of people in the documents has become more desirable. Although there is an automated system for installing such hyperlinks into judicial opinions, the system is not generally applicable to other types of names and documents, nor well suited to determine hyperlinks for names that might refer to two or more similarly named persons. Accordingly, the inventor devised systems, methods, and software that facilitate hyperlinking names in documents, regardless of type. One exemplary system includes a descriptor module and a linking module. The descriptor module develops descriptive patterns for selecting co-occurent document information that is useful in recognizing associations between names and professional classes. The linking module tags names in an input document, extracts co-occurent information using the descriptive patterns, and uses a Bayesian inference network that processes a (non-inverse-document-frequency) name-rarity score for each name along with the name and selected co-occurent document information to determine appropriate hyperlinks to other documents, such as entries in professional directories.

US Patent:

20090198678, Aug 6, 2009

Filed:

Dec 22, 2008

Appl. No.:

12/341913

Inventors:

Jack G. Conrad - Eagan MN, US
Christopher C. Dozier - Minneapolis MN, US
Sriharsha Veeramachaneni - St. Paul MN, US

International Classification:

G06F 17/30

US Classification:

707 5, 707E17014

Abstract:

To facilitate access to public records, the present inventors devised, among other things, an entity resolution system. The exemplary system includes master records database of 300 million entities, which is partitioned into multiple distinct portions. The exemplary system extracts entity information from input public records and constructs one or more blocking queries against specific portions of the master records database to identify one or more sets of candidate records. Feature vectors are defined for the candidate records and machine learning techniques, such as Support Vector Machine, are used to determine which of the candidate records from the master records database match the input public records. Candidate records that match are logically associated with public records, enabling ready access via direct or indirect queries.

US Patent:

20090222395, Sep 3, 2009

Filed:

Dec 22, 2008

Appl. No.:

12/341926

Inventors:

Marc Light - St. Paul MN, US
Frank Schilder - St. Paul MN, US
Ravi Kumar Kondadadi - Eagan MN, US
Christopher C. Dozier - Minneapolis MN, US
Wenhui Liao - Minneapolis MN, US
Sriharsha Veeramachaneni - St. Paul MN, US

International Classification:

G06N 5/02

US Classification:

706 47, 706 50, 706 54

Abstract:

For automated text processing, the inventors devised, among other things, an exemplary system that includes an entity tagger, an entity resolver, a text segment classifier, and a relationship extractor. The entity tagger receives an input text segment, and tags named entities with the segment as being a person, company, or place. The entity resolver accesses authority files, and associates the persons and companies named in the text segment with specific entries in the files. The text segment classifier determines whether the text segment includes a relationship event, such as job-change event or merger and acquisition event, and if an event is detected, the relationship extractor determines the event role of entities named in the segment. For example, the extractor determines for a merger and acquisition event, which named company was the acquirer and which was acquired.

US Patent:

20130191329, Jul 25, 2013

Filed:

Mar 30, 2012

Appl. No.:

13/435889

Inventors:

Christopher Dozier - Minneapolis MN, US
Souptik Datta - Jersey City NJ, US
Merine Thomas - Eagan MN, US

International Classification:

G06F 17/30

US Classification:

707609

Abstract:

A method includes receiving an entity record, wherein the entity record comprising at least one entity field element, and resolving the entity record to an authority record being associated with an initial authority file, wherein the authority record comprising at least one authority field element. The method further includes calculating a field element update measurement, the field element update measurement being associated with the at least one entity field element and the at least one authority field element and if the field element update measurement meets or exceeds a threshold, updating the authority record, or if the field element update measurement does not meet or exceed a threshold, not updating the authority record. The method further includes developing, in response to updating the authority record, an updated authority file associated with at least one updated authority record. The method also includes incorporating an additional authority record to the updated authority file.