Also available in PS and PDF
formats.
Okan Kolak
Computer Science
Department, University of Maryland, College Park, MD 20742
my_first_name_AT_umiacs.umd.edu • http://www.cs.umd.edu/users/okan/
Research Interests
Natural
language processing; in particular statistical modeling and methods,
resource acquisition and transfer using parallel corpora, machine
translation. Information retrieval.
Education
University of
Maryland, College Park, MD, USA
Ph.D.
candidate in Computer Science, December 2005 (expected)
Advisor:
Prof. Philip Resnik
University of
Maryland, College Park, MD, USA
M.S.
in Computer Science, May 2003, GPA: 4.0/4.0
Advisor:
Prof. Philip Resnik,
Bilkent
University, Ankara, Turkey
B.S.
in Computer Engineering and Information Science, May 1998, GPA:
3.55/4.0
Advisor:
Prof. Kemal Oflazer
Research Experience
University of
Maryland, College Park, MD, USA
Research
Assistant, Computational Linguistics and Information
Processing (CLIP) Lab, August 1999 – cont.
-
Designed a
generative probabilistic optical character recognition model (OCR),
developed parameter estimation methods for the model, and employed the
model for OCR post-processing, resulting in significant reduction in
recognition error rate. Evaluated the degradation caused by OCR errors
on NLP applications, and demonstrated that post-processing has a
positive impact.
-
Designed and
implemented Direct Projection Algorithm, which transfers dependency
parse trees on one side of a parallel corpus to the other using word
level alignments.
-
Worked on use
and evaluation of domain-tuned lexicons in statistical machine
translation systems.
-
Contributed
to various cross language information retrieval systems developed within
the CLIP Lab.
Microsoft,
Redmond, WA, USA
Software
Design Engineer Intern, Adaptive User Interface Group,
June – August 2000
-
Conducted an
initial feasibility study of speech based interface support for the
Mobile Controls, a unified development toolkit for creating adaptive Web
interfaces accessible by a diverse range of clients from computers to
cell phones. Designed and implemented a prototype that employed
Microsoft VoiceXML.
Computer and
Communications Research Lab, NEC USA, San Jose, CA, USA
Research
Assistant, June – August 1999
-
Worked on
several components of the NetTopix Focused Search Engine Project.
Introduced the concept of logical domains, which is used to cluster
search results and to generate multi-granular, topic-focused Web site
maps. Designed and implemented an algorithm for identifying and
extracting logical domains in a physical web domain. Developed the
full-text search component that allowed weighted term selection and
supported several ranking and presentation schemes.
Teaching Experience
University of
Maryland, College Park, MD, USA
Volunteer
Teaching Assistant, KNES147N SCUBA Diving, February 2001 – July
2004
University of
Maryland, College Park, MD, USA
Teaching
Assistant, CMSC723/LING645 Intro to Computational Linguistics,
February – June 2000
University of
Maryland, College Park, MD, USA
Teaching
Assistant, CMSC106 Introduction to C Programming, August 1998 –
June 1999
Bilkent
University Computer Club, Ankara, Turkey
Volunteer
Instructor, 1996 – 1998
-
Designed and
delivered various lectures ranging from basic computer skills to HTML.
Esenevler High
School, Ankara, Turkey
Volunteer
Instructor, September 1996 – June 1997
-
Taught basic
computer skills and user applications such as word-processing and
spreadsheets as part of the Computer Literacy Courses Program of the
Bilkent University Public Services.
Journal Publications
-
Rebecca Hwa,
Philip Resnik, Amy Weinberg, Clara Cabezas, and Okan Kolak.
“Bootstrapping Parsers via Syntactic Projection across Parallel Texts”,
In the Special Issue of the Journal of Natural Language Engineering
on Parallel Texts, Eds. Rada Mihalcea and Michel Simard. To appear.
-
Necip Fazil
Ayan, Wen-Syan Li, Okan Kolak, “Automating Extraction of Logical Domains
in a Web Site”, In International Journal of Data and Knowledge
Engineering, 43(2), Elsevier Science, pp. 179-205, November 2002.
Other Refereed Publications
-
Okan Kolak
and Philip Resnik, "OCR Post-Processing for Low Density Languages", In
Proceedings of the Human Language Technology Conference (HLT-EMNLP
2005), Vancouver, Canada, To appear.
-
Necip Fazil
Ayan, Bonnie Dorr, Okan Kolak, “Domain Tuning of Bilingual Lexicons for
MT”, In Proceedings of the Evaluation Workshop at the MT Summit IX,
New Orleans, Louisiana, USA, pp. 3-11, September 2003.
-
Okan Kolak,
William Byrne, Philip Resnik, “A Generative Probabilistic OCR Model for
NLP Applications”, In Proceedings of the Human Language Technology
Conference (HLT-NAACL 2003), Edmonton, Canada, May 2003.
-
Rebecca Hwa,
Philip Resnik, Amy Weinberg, and Okan Kolak, “Evaluating Translational
Correspondence using Annotation Projection”, In
Proceedings of the 40th Annual Meeting of the Association for
Computational Linguistics (ACL-02), Philadelphia, Pennsylvania, USA,
July 2002.
-
Okan Kolak
and Philip Resnik, “OCR Error Correction Using a Noisy Channel Model”,
In Proceedings of the Human Language Technology Conference (HLT 2002),
San Diego, California, USA, March 2002.
-
Wen-Syan Li,
Necip Fazil Ayan, Okan Kolak, Quoc Vu, Hajime Takano, and Hisashi
Shimamura, “Constructing Multi-Granular
and Topic-Focused Web Site Maps”, In Proceedings of 10th World Wide
Web Conference, Hong Kong, China, May 2001.
-
Wen-Syan Li,
Okan Kolak, Quoc Vu, and Hajime Takano, “Defining Logical Domains in a
Web Site”, In Proceedings of the 2000 ACM Hypertext Conference,
San Antonio, Texas, USA, May 2000.
-
Okan Kolak
and Wen-Syan Li, “On Ranking and Organizing Web Query Results”, In
Proceedings of the 1999 IEEE Knowledge and Data Engineering Exchange
Workshop (KDEX), Chicago, Illinois, USA, November 8, 1999.
Patents
-
Wen-Syan Li,
Okan Kolak, and Quoc Vu, “Method of Defining and Utilizing Logical
Domains to Partition and to Reorganize Physical Domains” US Patent
No. 6,647,381 issued on November 11, 2003, assigned to NEC USA,
Inc., Princeton, New Jersey, USA.
Software Experience
C++, C, C#,
Perl, Java, Pascal, Delphi, Lisp, Prolog, COBOL, SQL, CGI, AT&T FSM
Toolkit, UNIX, Linux, Windows.
Relevant Coursework
Natural
Language Processing, AI Planning, Neural Modeling, Programming Language
Implementation—Implementing Java, Database Systems Implementation, Data
Structures, Computer Networks, Compiler Design, Distributed Systems,
Computational Geometry, Computer Graphics, High Performance Computing,
Program Verification, Software Engineering.
Honors
-
1998 Ranked 2nd in Computer Eng. and Information Sci. Dpt.,
Bilkent University
-
1993-98 Full Scholarship, including tuition, stipend, and
accommodation, Bilkent University
-
1997 Certificate of Acknowledgement, Bilkent University
Public Services
-
1994-98 Dean’s High Honor List, Bilkent University
-
1993 Scholarship, for undergraduate study abroad, Turkish
Republic Ministry of Education
Personal
-
Permanent U.S. Resident
-
SCUBA certifications: Advanced Diver, Nitrox Diver,
Rescue Diver -
Class B Commercial Driver License with Air Brake and
Passenger endorsements
References
|