Engineer Software
Resume:
Alireza Farhangfar
**** ******** *****, ***. ****
Sunnyvale, California 94085
Tel: 650-***-****
Email: ***********@*****.***
Citizenship: Canadian
(eligible to work in US with TN visa and currently residing in the bay area)
SUMMARY OF SKILLS
Solid background in machine learning and data mining techniques including classification,
regression, statistical modeling, active learning, missing data imputation and recommender
systems
Experience with data analysis and data visualization on large real data
Extensive experience with programming languages C/C++, Java, Python
Extensive experience with numerical analysis tool MATLAB, machine learning software
Weka, relation database MYSQL
Publications in several top conferences and journals in machine learning and one patent
pending
EDUCATION
Ph.D. in Computer Science (Machine Learning) Mar. 2013
University of Alberta, Edmonton, Canada
Advisors: Russ Greiner, Csaba Szepesvari
M.Sc. in Computer Engineering Dec. 2004
University of Alberta, Edmonton, Canada
M.Sc. in Electrical Engineering Sep. 2002
University of Tehran, Iran
B.Sc. Degree in Electrical Engineering Sep. 2000
University of Tehran, Iran
REASERCH EXPERIANCE
Research Assistant: Department of Computing Science, University of Alberta, Sep. 2006-Mar.2013
Developed various active learning algorithms for classification tasks
Introduced importance sampling active learning algorithm (ISAL)
o ISAL a sample and label efficient algorithm useful for applications where an
efficient data collection procedure is required
o It sequentially provides a distribution that puts large weight on instances whose
labels are uncertain, then requests the label of an instance drawn from that
distribution
Developed an algorithm for actively learning the classification model for structured data
such as images
o Given a large number of unsegmented images, and access to a human exp ert who
can segment a given image, the proposed active learner decides which images to
query, to quickly produce a segmenter that is accurate over this distribution of
images.
o The proposed active learner produces an effective segmenter using a few
segmented images on real world datasets
Research Assistant: Department of Electrical and Computer Engineering, University of Alberta, Sep.
2002–Jun. 2005
Developed a novel framework for imputation of missing values in datasets
Studied the impact of imputation of missing values on classification error of various
classifiers including decision trees and K-nearest-neighbor
WORK EXPERIANCE
May 2010–Sep. 2010
Research Engineer Intern in Machine Learning
Robert Bosch Research and Technology Center
Palo Alto, California, USA
Developed a system to match patients with a telemedicine system
o Our system identifies the patients that may benefit from the telemedicine system
in order to reduce their rate of hospitalization
Performed statistical analysis on large medical data to showcase the benefits of using
telemedicine system for the chronic patients
Presented the work at the Healthcare Division of Robert Bosch LLC
Nov. 2005–Sep. 2006
Software Engineer
Alberta Ingenuity Center for Machine Learning (AICML),
Edmonton, Alberta, Canada
Worked on the brain tumor analysis project to apply machine learning and computer
vision techniques to MRI of human brain
Worked with a team of researchers from computer science and oncology departments to
develop a state of the art system to segment the MRI scans and identify the location of the
tumor inside human brain
COMPUTER SKILLS
Extensive experience in programming with C/C++, Java, Phyton
Extensive experience with relational databases using MYSQL, numerical analysis tool MATLAB,
and machine learning software Weka
PUBLICATIONS
Patent Application
1. Srinivasan S., Farhangfar, A., Methods and Systems for Selection of Patients to Receive
a Medical Device, US patent application 13296788, filed Nov. 2011.
Journal and Conference papers
1. Farhangfar A., Greiner G., Szepesvári C., Importance Sampling Active Learning, to be
submitted to NIPS, 2013.
2. Farhangfar A., Greiner G., Szepesvári C., Learning to Segment from a Few Well-
Selected Training Images, International Conference on Machine Learning (ICML), June
2009.
3. Farhangfar, A., Kurgan, L., and Dy, J., Impact of Imputation of Missing Values on
Classification Error for Discrete Data, Journal of Pattern Recognition, Dec. 2008,
Volume 41, Issue 12, pp. 3692-3705.
4. Farhangfar A., Greiner G., Zinkevich, M., A Fast Way to Produce Near-Optimal Fixed-
Depth Decision Trees, The Tenth International Symposium on Artificial Intelligence and
Mathematics (ISAIM2008), Fort Lauderdale, Florida, Jan. 2008.
5. Farhangfar, A., Kurgan, L., and Pedrycz, W., Novel Framework for Imputation of
Missing Values in Databases, IEEE Transactions on Systems, Man and Cybernetics, Part
A: Systems and Humans, 37 (5), 2007, pp. 692-709.
6. Farhangfar, A., Kurgan, L., and Pedrycz, W., Experimental Analysis of Methods for
Handling Missing Values in Databases, Intelligent Computing: Theory and Applications
II Conference, held in conjunction with the SPIE Defense and Security Symposium
(formerly AeroSense), Orlando, FL, 2004.
7. Farhangfar, A., Afsharnia, S., and Sajjadi, S.J., Power Flow Control and Loss
Minimization with Unified Power Flow Controller (UPFC), IEEE Canadian Conference
on Electrical and Computer Engineering.(CCECE), 2004.
8. Farhangfar, A., Kurgan, L., and Pedrycz, W., Novel Method for Handling Missing
Values in Databases Based on Mean Pre-Imputation, Confidence Intervals and Boosting,
MITACS 5th annual conference, June 2004.
Invited Talks
1. Active learning on structured data, Yahoo! Research, Sunnyvale, CA, Nov. 2011.
2. Introduction to collaborate filtering, University of Alberta, Edmonton, AB, Mar. 2013.
AWARDS
Informatics Circle of Research Excellence (iCORE) ICT Scholarship, 2008 –2012
NSERC Postgraduate Scholarship (PGSD3), 2006-2009
Alberta Ingenuity graduate Scholarship, 2007-2008
iCORE Graduate Student Scholarship, 2006-2008
Walter H. Johns Graduate Fellowship, 2006-2009
Faculty of Science Graduate Entrance Scholarship, 2006
Contact this candidate