Search

Gabriela C Surpi

from San Diego, CA
Age ~58
Get Report
Also known as:
Gabriela Clara Surpi
Phone and address:
8215 Harton Pl, San Diego, CA 92123
(858) 292-4224
Connected to:
Alexander Surpin, 42Donn C Surpless, 79John M Surplus, 68Bob A Surprenant, 59Careid D Surprenant, 47Carolyn E Surprenant, 67Chris Surprenant, 53Dawn D Surprenant, 59Erica R Surprenant, 42Gerard J Surprenant, 82

Gabriela Surpi Phones & Addresses

  • 8215 Harton Pl, San Diego, CA 92123 (858) 292-4224
  • 35Th St, San Diego, CA 92104
  • Pasadena, CA

Resumes

Resumes

Gabriela Surpi Photo 1

Senior Machine Learning Scientist

View page
Location:
San Diego, CA
Industry:
Information Technology And Services
Work:
Amazon
Senior Machine Learning Scientist
Lytx, Inc. Jul 2014 - Oct 2017
Principal Scientist and Director of Advanced Technology
Lytx, Inc. Oct 2012 - Jun 2014
Senior Scientist
Fico Dec 2003 - Apr 2011
Lead Scientist
Hnc Dec 2001 - Nov 2003
Analytic Staff Scientist
Education:
University of Buenos Aires Nov 1997
University of Buenos Aires 1992 - 1997
Doctorates, Doctor of Philosophy, Physics University of Hamburg Aug 1996
University of Buenos Aires Feb 1996
University of Buenos Aires 1995
University of Buenos Aires May 1992
Master of Science, Masters
Skills:
Predictive Analytics
Data Mining
Statistical Modeling
Business Intelligence
Machine Learning
Algorithms
E Commerce
Product Management
Predictive Modeling
Artificial Intelligence
Logistic Regression
Research
Neural Networks
Credit Scoring
Analysis
Natural Language Processing
Analytics
Data Analysis
Text Mining
Modeling
Statistics
Python
Sas
Business Analytics
Crm
Perl
Product Development
R
Pattern Recognition
Gabriela Surpi Photo 2

Gabriela Surpi

View page
Location:
San Diego, CA
Industry:
Financial Services

Publications

Us Patents

System And Method For Building A Predictive Score Without Model Training

View page
US Patent:
20110131164, Jun 2, 2011
Filed:
Dec 2, 2009
Appl. No.:
12/629712
Inventors:
GABRIELA SURPI - San Diego CA, US
International Classification:
G06N 5/02
G06F 15/18
G06N 3/12
US Classification:
706 13, 706 21, 706 46, 706 45
Abstract:
A system and method for building a predictive score without model training are disclosed. A set of predictive variables is defined based on raw data fields generated from raw data from one or more sources and domain knowledge. The raw data includes a historical set of transactions previously generated by one or more raw data sources. An scaled relative risk table to describe each predictive variable of the set of predictive variables is generated. The set of predictive variables is combined based on their associated relative risk tables to generate a predictive score for a future set of transactions.

System And Method For Building A Predictive Score Without Model Training

View page
US Patent:
20130232100, Sep 5, 2013
Filed:
Apr 15, 2013
Appl. No.:
13/863292
Inventors:
Gabriela Surpi - San Diego CA, US
International Classification:
G06N 5/02
US Classification:
706 46
Abstract:
A system and method for building a predictive score without model training are disclosed. A set of predictive variables is defined based on raw data fields generated from raw data from one or more sources and domain knowledge. The raw data includes a historical set of transactions previously generated by one or more raw data sources. An scaled relative risk table to describe each predictive variable of the set of predictive variables is generated. The set of predictive variables is combined based on their associated relative risk tables to generate a predictive score for a future set of transactions.

Detecting Risky Driving With Machine Vision

View page
US Patent:
20170061222, Mar 2, 2017
Filed:
Aug 31, 2015
Appl. No.:
14/841376
Inventors:
- San Diego CA, US
Gabriela Surpi - San Diego CA, US
International Classification:
G06K 9/00
G07C 5/00
G07C 5/06
Abstract:
A system for driver risk determination using machine vision includes an interface and one or more processors. The interface is to receive one or more video data streams. The one or more processors is/are to determine, whether a driving behavior risk type appears in one of the one or more video data streams; and in the event that the driving behavior risk type appears in the one of the one or more video data streams, indicate the driver behavior risk type.

Back-End Event Risk Assessment With Historical Coaching Profiles

View page
US Patent:
20160314707, Oct 27, 2016
Filed:
Jul 1, 2016
Appl. No.:
15/201198
Inventors:
- San Diego CA, US
Gabriela Surpi - San Diego CA, US
International Classification:
G09B 19/16
G09B 5/02
B60W 40/09
G07C 5/08
H04L 29/08
G06F 17/30
Abstract:
A system for transferring data comprises an input interface and a processor. The input interface is configured to receive event information from a vehicle event recorder including an event type for each event associated with the event information. The processor is configured to determine a historical coachable profile for event types associated with the event information and select a group of events to transfer based at least in part on the historical coachable profile for the event types.

Dynamic Uploading Protocol

View page
US Patent:
20150310676, Oct 29, 2015
Filed:
Jun 30, 2015
Appl. No.:
14/755948
Inventors:
- San Diego CA, US
Larry Richardson - San Diego CA, US
Syrus C. Nemat-Nasser - San Diego CA, US
Gabriela Surpi - San Diego CA, US
International Classification:
G07C 5/00
G06N 7/00
H04L 29/08
Abstract:
A dynamic uploading protocol comprises an input interface configured to receive a manifest comprising a plurality of events which may be uploaded; wherein the manifest additionally comprises sensor information relating to each of the plurality of events. The system for a dynamic uploading protocol additionally comprises a processor configured to determine whether to upload additional information about each event, wherein determining whether to upload additional information about each event is based at least in part on the sensor information and contextual information. The system for a dynamic uploading protocol additionally comprises an output interface configured to request the additional information. The system for a dynamic uploading protocol additionally comprises a memory coupled to the processor and configured to provide the processor with instructions.

Dynamic Uploading Protocol

View page
US Patent:
20150088335, Mar 26, 2015
Filed:
Sep 26, 2013
Appl. No.:
14/037815
Inventors:
- San Diego CA, US
Larry Richardson - San Diego CA, US
Syrus C. Nemat-Nasser - San Diego CA, US
Gabriela Surpi - San Diego CA, US
Assignee:
Lytx, Inc. - San Diego CA
International Classification:
G07C 5/00
US Classification:
701 1
Abstract:
A dynamic uploading protocol comprises an input interface configured to receive a manifest comprising a plurality of events which may be uploaded; wherein the manifest additionally comprises sensor information relating to each of the plurality of events. The system for a dynamic uploading protocol additionally comprises a processor configured to determine whether to upload additional information about each event, wherein determining whether to upload additional information about each event is based at least in part on the sensor information and contextual information. The system for a dynamic uploading protocol additionally comprises an output interface configured to request the additional information. The system for a dynamic uploading protocol additionally comprises a memory coupled to the processor and configured to provide the processor with instructions.
Control profile
Gabriela C Surpi from San Diego, CA, age ~58 Get Report