Rayid Ghani

US Patent:

7904378, Mar 8, 2011

Filed:

Jun 15, 2010

Appl. No.:

12/816079

Inventors:

Rayid Ghani - Chicago IL, US
Hillery D. Simmons - Chicago IL, US

Assignee:

Accenture Global Services Limited

International Classification:

G06Q 40/00

US Classification:

705 37

Abstract:

An auction result prediction system predicts auction results. The system may determine item, seller, or auction characteristics from prior or pending auctions. The system also obtains item characteristics of an item for which a result prediction is sought, either by a buyer or by a seller. A price predictor in the system accepts the auction and item characteristics and predicts an auction result based on the characteristics. The system also determines insurance parameters for insuring online auctions, and the insurance parameters may be based on predicted auction results. An insurance policy reflecting the insurance parameters may be offered to an online auction buyer, seller, or other market participant. The insurance policy may insure, for example, that an item for sale will obtain at least a price specified by the insurance policy.

US Patent:

7945473, May 17, 2011

Filed:

Feb 28, 2005

Appl. No.:

11/069472

Inventors:

Andrew E. Fano - Lincolnshire IL, US
Chad M. Cumby - Chicago IL, US
Rayid Ghani - Evanston IL, US
Marko Krema - Evanston IL, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06Q 30/00

US Classification:

705 141, 705 10

Abstract:

A method and system for using individualized customer models when operating a retail establishment is provided. The individualized customer models may be generated using statistical analysis of transaction data for the customer, thereby generating sub-models and attributes tailored to customer. The individualized customer models may be used in any aspect of a retail establishment's operations, ranging from supply chain management issues, inventory control, promotion planning (such as selecting parameters for a promotion or simulating results of a promotion), to customer interaction (such as providing a shopping list or providing individualized promotions).

US Patent:

7970767, Jun 28, 2011

Filed:

Apr 30, 2007

Appl. No.:

11/742244

Inventors:

Katharina Probst - Dyer IN, US
Rayid Ghani - Chicago IL, US
Andrew E. Fano - Lincolnshire IL, US
Marko Krema - Evanston IL, US
Yan Liu - New York City NY, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06F 17/30

US Classification:

707739, 707777, 707796, 707803

Abstract:

One or more classification algorithms are applied to at least one natural language document in order to extract both attributes and values of a given product. Supervised classification algorithms, semi-supervised classification algorithms, unsupervised classification algorithms or combinations of such classification algorithms may be employed for this purpose. The at least one natural language document may be obtained via a public communication network. Two or more attributes (or two or more values) thus identified may be merged to form one or more attribute phrases or value phrases. Once attributes and values have been extracted in this manner, association or linking operations may be performed to establish attribute-value pairs that are descriptive of the product. In a presently preferred embodiment, an (unsupervised) algorithm is used to generate seed attributes and values which can then support a supervised or semi-supervised classification algorithm.

US Patent:

8027941, Sep 27, 2011

Filed:

Sep 14, 2007

Appl. No.:

11/855493

Inventors:

Katharina Probst - Dyer IN, US
Rayid Ghani - Chicago IL, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06F 15/18
G06E 1/00

US Classification:

706 14, 706 15, 706 20

Abstract:

A classification algorithm is separated into one or more input-invariant parts and one or more input-dependent classification parts. The input-invariant parts of the classification algorithm capture the underlying and unchanging relationships between the plurality of data elements being operated upon by the classification algorithm, whereas the one or more classification parts embody the probabilistic labeling of the data elements according to the various classifications. For any given iteration, a user's input is used to modify at least one classification part of the algorithm. Recalculated classification parts (i. e. , updated classification results) are determined based on computationally simple combinations of the one or more modified classification parts and the one or more input-invariant parts. Preferably, a graphical user interface is used to solicit user input. In this manner, wait times between user feedback iterations can be dramatically reduced, thereby making application of active learning to classification tasks a practical reality.

US Patent:

8117199, Feb 14, 2012

Filed:

Feb 23, 2010

Appl. No.:

12/710832

Inventors:

Rayid Ghani - Evanston IL, US
Andrew E. Fano - Lincolnshire IL, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06F 7/00
G06F 17/30
G06F 17/00

US Classification:

707734, 707749, 707944, 705 733, 705 267

Abstract:

Relative to a given product or products, one or more attributes and, for each attribute, a plurality of possible attribute values, are defined. For a given product and attribute, one or more descriptions of the product are obtained and analyzed to determine the correspondence of the description(s), and hence the product itself, to each of the plurality of possible attribute values. In one embodiment, this analysis is based on previously-labeled training data. A knowledge base can be populated with information identifying the products and their correspondence to the plurality of possible attribute values for each attribute. This technique may be used to develop a profile of an entity, which in turn may be used to develop appropriate marketing messages or recommendations for other products.

US Patent:

8417653, Apr 9, 2013

Filed:

Aug 24, 2011

Appl. No.:

13/216840

Inventors:

Katharina Probst - Dyer IN, US
Rayid Ghani - Chicago IL, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06F 15/18

US Classification:

706 14

Abstract:

A classification algorithm is separated into one or more input-invariant parts and one or more input-dependent classification parts. Classifiable electronic data is obtained via a communication network. Using the classification algorithm, classifications of a plurality of data elements in the classifiable data are identified, where the at least one classification part incorporates user input concerning classification of at least one data element of the plurality of data elements.

US Patent:

8504492, Aug 6, 2013

Filed:

Jan 10, 2011

Appl. No.:

12/987505

Inventors:

Rayid Ghani - Chicago IL, US
Chad Cumby - Chicago IL, US
Marko Krema - Evanston IL, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06F 15/18

US Classification:

706 12

Abstract:

A body of text comprises a plurality of unknown attributes and a plurality of unknown values. A first classification sub-component labels a first portion of the plurality of unknown values as a first set of values, whereas a second classification sub-component labels a portion of the plurality of unknown attributes as a set of attributes and a second portion of the plurality of unknown values as a second set of values. Learning models implemented by the first and second classification subcomponents are updated based on the set of attributes and the first and second set of values. The first classification sub-component implements at least one supervised classification technique, whereas the second classification sub-component implements an unsupervised and/or semi-supervised classification technique. Active learning may be employed to provide at least one of a corrected attribute and/or corrected value that may be used to update the learning models.

US Patent:

8521745, Aug 27, 2013

Filed:

Jun 13, 2011

Appl. No.:

13/158678

Inventors:

Katharina Probst - Dyer IN, US
Rayid Ghani - Chicago IL, US
Andrew E. Fano - Lincolnshire IL, US
Marko Krema - Evanston IL, US
Yan Liu - Elmsford NY, US

Assignee:

Accenture Global Services Limited - Dublin

International Classification:

G06F 17/30

US Classification:

707739

Abstract:

One or more classification algorithms are applied to at least one natural language document in order to extract both attributes and values of a given product. Supervised classification algorithms, semi-supervised classification algorithms, unsupervised classification algorithms or combinations of such classification algorithms may be employed for this purpose. The at least one natural language document may be obtained via a public communication network. Two or more attributes (or two or more values) thus identified may be merged to form one or more attribute phrases or value phrases. Once attributes and values have been extracted in this manner, association or linking operations may be performed to establish attribute-value pairs that are descriptive of the product. In a presently preferred embodiment, an (unsupervised) algorithm is used to generate seed attributes and values which can then support a supervised or semi-supervised classification algorithm.