Mohammad Shahed Sorower

th

**** ** ** **., *** # G, Corvallis, OR 97330 E- mail: abpous@r.postjobfree.com

Phone: 901-216- 6168 W eb: http://people.oregonstate.edu/~sorowerm/

EDUCATION

PhD in C omputer Science Sept 2009 - P resent

Oregon State University. Corvallis, Oregon.

MS in Electrical and Computer Engineering, GPA : 4.0/4.0 D ec 2008

The University of Memphis, Memphis, Tennessee.

BS in Computer Science and Engineering, G PA : 3.80/4.0 Nov 2004

Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh.

RESEARCH EXPERIENCE

Oregon State University, Dept. of Computer Science 09/09-Present

Graduate Research Assistant, Research Focus: Machine Learning, NLP, HCI

A G oogle funded (ended Summer 12) project TaskTracer, which is a part of our effort to build

intelligent desktop assistants; a nice computer program that organizes all the task information in

one place rather than distributed across local and network files systems, emails, websites,

calendars, contacts etc.

A D ARPA funded project (ended Summer 12) titled a s "Machine Reading", in collaboration with

Carnegie Mellon Univ. and Univ. of Washington, ISI and BBN technologies a nd Cycorp. Goal for my

work is to learn inference rules from incomplete texts using undirected graphical model (Markov

Logic Networks).

The University of Memphis, Dept. of Electrical and Computer Engg. 08/08- Present

Herff Fellow, Research Focus: Intelligent Systems

Human Computer Interaction/Natural Language Processing: Worked towards building a framework

to recognize human emotion using multimodal fusion; audio, video and text features.

The University of Memphis, FedEx Institute of Technology 10/06-10/07, 10/07- 05/08

Senior Investigator, System Testing Excellence Program (STEP)

R esearch Focus: Towards p roviding a unified solution for software testing using machine learning.

The University of Memphis, Dept. of Electrical and Computer Engg. 08/06- 08/08

Computer Vision, Pattern and Image Analysis Lab, Graduate Research Assistant

R esearch Focus: Pattern Recognition, Information Retrieval and Semantic Analysis.

Bangladesh University of Engineering and Technology, Dept. of CSE

Undergraduate Thesis: Pattern Recognition Using Neural Networks; Mathematical Polynomials and

Radial coding to develop a High P erformance Invariant Character Recognition System.

PROFESSIONAL EXPERIENCE

Microsoft Research, Research Intern 06/12- 09/12

Semantic Parsing of natural language texts is the bottleneck for a variety of NLP problems due to

"knowledge capacity" and "computability" of the existing techniques. In our work, we tried to work

on a very general unsupervised semantic parsing technique b y inducing "unsupervised supervision"

into the system.

IBM, Application & Integration Middleware S/W, Performance Engg. Co- Op 05/08-08/08

Security Automation ( i.e. administration, application and resource sharing, encryption and audit)

STAF for W ebsphere Application Server performance evaluation using STAX, Java, Jython and Jacl.

AutoZone Inc., Summer IT (Security) Intern 05/07- 08/07

Automatic secure (i.e. PGP Encryption) file transfer (i.e. scripting in Expect, Perl)

File integrity checkers (i.e. Samhain)

TM International (Bangladesh) Ltd.(TMIB), System Analyst 03/05- 08/06

Firewall Security, VPN Operations, VoIP setup and billing (RADIUS, MYSQL/Oracle, PHP, JS).

Information and IP Network Security policy development to achieve ISO27001

SELECTED PUBLICATIONS AND POSTER

Mohammad Shahed Sorower, Thomas G. Dietterich, Janardhan Rao Doppa, Walker Orr, Xiaoli

Fern and Prasad Tadepalli. Inverting Grice's Maxims to Learn Rules from Natural Language

Extractions. Proceedings of Advances in Neural Information Processing Systems (NIPS-2011),

Granada, Spain, Dec 12- 15, 2011.

Janardhan Rao Doppa, Mohammad Shahed Sorower, Mohammad Nasresfahani, Jed Irvine,

Walker Orr, Thomas G. Dietterich, Xiaoli Fern and Prasad Tadepalli. Learning Rules from

Incomplete Examples via Implicit Mention Models. Journal of Machine Learning Research (JMLR)

Proceedings Track, volume 20 (ACML-2011).

Sorower, M. S., Dietterich, T. G., Doppa, J. R., Tadepalli, P., Fern, X. (2011). Learning Rules from

Incomplete Examples via a Probabilistic Mention Model. IJCAI 2011 Workshop on Learning by

Reading and its Applications in Intelligent Question-Answering, Barcelona, Spain.

Doppa, J. R., NasrEsfahani, M., Sorower, M. S., Irvine, J. Dietterich, T. G., Fern, X., Tadepalli, P.

(2011). Learning Rules from Incomplete Examples via Observation Models. IJCAI 2011 Workshop

on Learning by Reading and its Applications in Intelligent Question- Answering, Barcelona, Spain.

Mohammad S Sorower, Thomas G Dietterich. Mention Model for Learning Rules from Incomplete

Examples. Poster in Student Summit, jointly organized with Machine Reading Phase 3 kick-off

meeting, Seattle, WA. April 5-8, 2011.

Janardhan Rao Doppa, Mohammad Nasresfahani, Mohammad S. Sorower, Thomas G. Dietterich,

Xiaoli Fern and Prasad Tadepalli, Towards Learning Rules from Natural Texts, In Proceedings of

Workshop on Formalisms and Methodologies in Learning by Reading (NAACL-2010), Los Angles,

California.

Mohammed Yeasin, Haritha Malempati, Ramin Homayouni and Mohammad Shahed Sorower, A

systematic study on latent semantic analysis model parameters for mining biomedical literature,

UT- ORNL- KBRIN Bioinformatics Summit, Pikeville, TN, USA. 20 22 March 2009 and BMC

Bioinformatics 2009, 10(Suppl 7):A6 doi:10.1186/1471- 2105- 10-S7-A6.

Mohammad S Sorower, Mohammed Yeasin, Role of Intonation for Modeling Dialog Acts from the

Transcription of Utterances, Poster for Annual Meeting of the Society for Text and Discourse,

Rotterdam, Netherlands, July 26-28, 2009.

D.J. Russomanno, M. Yeasin, E. Jacobs, M. Smith, and M.S. Sorower Sparse Detector Sensor:

Profiling Experiments for Broad-Scale Classification, Proceedings SPIE-Defense and Security

Symposium: Unattended Ground, Sea, and Air Sensor Technologies and Applications X, vol. 6963,

2008, Orlando, Florida, CD. 11 pages.

M. Yeasin, D.J. Russomanno, M. S. Sorower, M.E. Smith and J.S. Shaik, Robust Classification of

Objects using a Sparse Detector Sensors, The 2008 International Conference on Machine Learning;

Models, Technologies and Applications (MLMTA'08), July 14- 17, 2008, Monte Carlo Resort, Las

Vegas, Nevada, USA.

M. E. Hoque, M. S. Sorower, M. Yeasin, M. M Louwerse, What Speech Tells Us About Discourse:

The Role of Prosodic and Discourse Features in Dialogue Act Classification, The 2007 International

Joint Conference on Neural Networks (IJCNN), Aug 12-17, 2007, Orlando, Florida, USA.

Mohammad S. Sorower, Mohammed Yeasin, Robust Classification of Dialog Acts from the

Transcription of Utterances, First IEEE International Conference on Semantic Computing (ICSC),

Sept 17- 19, 2007, Irvine, California, USA.

SELECTED DEVELOPMENT EXPERIENCE

Learning Based Reconstruction of Video from FACS in sync with Audio for Videophone, using Java.

Search Engine w ith common features for www.memphis.edu d omain using Perl and Perl CGI.

Content Based Music Retrieval System based on music genre domain using Perl and Perl CGI.

COMPUTER EXPERTISE

Programming Language: C, C++, C#, VB, JAVA, Matlab, Prolog, Assembly(8086)

Scripting Language: Shell Script, Perl, Expect, Python, JavaScript, PHP, HTML

Tools and Technology: OpenCV, .NET, OpenGL, UML

RDBMS: Oracle 8i, PL/SQL, MS SQL, MySQL

RELATED COURSEWORKS

Graduate Undergraduate

Artificial/Computational Intelligence Algorithms/Data Structures, Databases

Data Mining, N eural Networks Software Development/Engineering

Machine Learning, NLP/IR Calculus, Linear Algebra/Numerical Methods

PROFESSIONAL TRAINING

Intermediate System Administration for Solaris 9 Operating Environment, SA239.

Advanced System Administration for Solaris 9 Operating Environment, SA299.

Cisco Secured PIX Firewall Advanced, C SPFA.

HONORS

Herff Fellowship, Herff College of Engineering, The University of Memphis, 2008.

University Grant Commission (UGC), Dhaka, Bangladesh Scholarship, 2003.

Merit Scholarship, BUET, 2001-2002.

REFERENCES Available upon request

Contact this candidate