nsharifr@cs.cmu.edu Office: 6607 Gates Hillman Center, Carnegie Mellon University.

Narges Sharif-Razavian

 

Hi, and welcome to my webpage. I’m a fourth year PhD student in Language Technologies Institute, and I’m working with my advisor, Prof. Christopher James Langmead, on Graphical Models applied to protein structure modeling. We are currently working on Estimation and Inference for continuous multivariate models of backbone torsion angles, including nonparametric graphical models (embedded in Reproducing Kernel Hilbert Space, RKHS), time-varying multivariate Gaussian graphicl model, and sparse von-Mises graphical model. The next step is the analysis of these graphical models, in particular aiming at improving drug design application.

 

You can find my resume here: PDF

My Research Interests

Graphical Models: Kernel density estimation for graphical models, Parameter Estimation and Inference on Continuous Multivariate Models, Network Analysis, Applications: Protein Structure Prediction, Structural Biology, Drug Design, Gene Regulation Network Estimation, Genome-wide Association Studies,

 

Relevant Courses:

10-725 Optimization

15-879 Algorithms for Computational Structural Biology

10-810 Computational Genomics

11-762 Languages and Statistics II

Auditing: Physiology, Modern Biology, Bio-Chemistry

 

Teaching Assistantship

I have previously TA-ed the Machine Learning course, taught by Prof. Roni Rosenfeld in Spring 2011.

I am currently TA-ing the Algorithms for NLP Course 11-711, taught by Prof. Alon Lavie, Prof. Noah Smith, and Prof. Robert Frederking.

 

Publication and Technical Reports

• Narges Razavian,"Continuous Graphical Models for Static and Dynamic Distributions: Application to Structural Biology" PhD thesis proposal, December 2012
• Narges Razavian, Christopher J. Langmead, "Comparing Structure Learning Methods for RKHS Embeddings of Protein Structures", accepted in NIPS workshop on the Confluence between Kernel Methods and Graphical Models, December 2012
• Narges Razavian, Hetu Kamisetty, Christopher J. Langmead, Learning generative models of molecular dynamics, Tenth Asia Pacific Bioinformatics Conference (APBC 2012), also in BMC Genomics 2012
• Narges Razavian, Hetu Kamisetty, Christopher Langmead, “Expectation Propagation for von-Mises Graphical Models”, NIPS Workshop, Machine Learning in Computational Biology, December 2012
• Narges Razavian, Hetu Kamisetty, Christopher Langmead, “ The von Mises Graphical Model: Structure Learning”, NIPS Workshop on Machine Learning in Computational Biology, December 2011 
• Narges Razavian, Subhodeep Moitra, Hetu Kamisetty, Arvind Ramanathan, Christopher J. Langmead, “Time-Varying Gaussian Graphical Models of Molecular Dynamics Data” Proceedings of 3DSIG 2010 Structural Bioinformatics and Computational Biophysics, Boston, MA. July 9-10, 2010.
• Narges Razavian, Selen Uguroglu, Andreas Zollmann. “Species Selection for Phylogeny-Based Motif Detection”, Computational Genomics Technical Report, June 2009
• Narges Razavian, Andreas Zollmann. “An Overview of Nonparametric Bayesian Models and Applications to Natural Language Processing”, Languages and Statistics II project report, January 2009
• Narges Razavian, Stephan Vogel,“Fixed Length Word Suffix as New Factors in Factored Statistical Machine Translation”, ACL 2010, also presented in LTI Student Research Symposium, September 2010
• Narges Razavian, Stephan Vogel,“The Web as a Platform to Build Machine Translation Resources”, International Workshop on Intercultural Collaboration(IWIC2008), Stanford, USA February 2009
• Narges Razavian, Fattaneh Taghiyareh,“Embedding Corporate Blogging System in CRM Solutions” Information Technology: New Generations (ITNG), Las Vegas, USA, April 2008.
• Mostafa Keikha, Narges Razavian, Farhad Oroumchian, Hassan Seyedrazi,“Document Representation and Quality of Text: An Analysis”, Survey of Text Mining: Clustering, Classification, and Retrieval, Second Edition, Chapter 12, pp219-232,Springer London, ISBN 978-1-84800-046-9, July 2007.

Fun little projects:

In my Social Networks class project we developed a Facebook Application called Tetris For Research. It has about ~1300 monthly active users! Make sure to check it out Here

If you're a Farsi speaker, download my first ever (Free!) iphone app, Khayam Poems (in Farsi) from the app store, and enjoy some mind refreshing poetry.

 

Personal

In my free time I like to draw and paint (I’ve taken CMU’s figure drawing class and attend the weekly Sunday drawing session in Fine arts building), read books, and watch documentaries and movies. I travel a lot, and like to try new things.

Useful Links

I'm just beginning to gather a list of links I think might be useful one way or the other. The list is short now but hopefully I'll add more and more stuff to it as time goes on. If you found a particular link helpful drop me a line! 

• A nice collection of Machine Learning lectures: http://videolectures.net/Top/Computer_Science/Machine_Learning/
• A very nice course on Graphical Models, taught by Prof. Eric Xing. http://www.cs.cmu.edu/~epxing/Class/10708/
• A Farsi tagged corpus of size 2.6 million words, of daily news and common texts. The corpus is manually POS tagged and documents are categorized into 4300 different subjects: http://ece.ut.ac.ir/dbrg/bijankhan/
• A number of translated poems of Rumi, the great Persian poet of 13th Century. http://www.rumionfire.com/shams/index.htm
• Some cool data sets for your research: Stanford Large Network Dataset Collection, Youtube and Flickr and Digg data set , A large website full of data(Specially Twitter dataset), and lots of valuable Language(LDC) dataset
• A tutorial on Belief Propagation on Factor graphs PDF slides
• How to get your own Twitter data
• Google Map Mashup Page, mappying Iranian folk songs and cultural material to the city and location that they're related to.
• Something I've been looking for, for a while, and is not easy to find: Divan Hafez in Farsi (Hafez poems in farsi).
• Also, since many new comers have questions about coming to Pittsburgh, and I've repeated the same thing several times, I'm putting it in here: a guide to Pittsburgh so that new comers can get their answers while I can save some time... :)

Favorite Quotations

"Let the beauty of what you love be what you do". -- Rumi
"The way you think will create the reality for you" --Oprah Winfrey