• HOME
• RESEARCH
• TEACHING
• CV
• SCHEDULE
• LINKS

Research Interests:

• Geometric data analysis. Computational topology. Optimization. Numerical linear algebra. Analysis of high-dimensional data sets. Machine learning and pattern recognition. Hyperspectral imagery.

Publications:

• Sofya Chepushtanova, Elin Farnell, Eric Kehoe, Michael Kirby, and Henry Kvinge.
  Book chapter Dimensionalty Reduction in Data Science for Mathematicians.
  [Book homepage]
  Editor Nathan Carter, Chapman & Hall/CRC, New York, 2020.


• Sofya Chepushtanova and Michael Kirby.
  Sparse Grassmannian Embeddings for Hyperspectral Data Representation and Classification.
  [Publisher website]
  in IEEE Geoscience and Remote Sensing Letters, 2017, vol.14(3), pp.434 - 438,
  DOI: 10.1109/LGRS.2017.2648514
  


• Henry Adams, Sofya Chepushtanova, Tegan Emerson, Eric Hanson, Michael Kirby, Francis Motta, Rachel Neville, Chris Peterson, Patrick Shipman, and Lori Ziegelmeier.
  Persistence Images: A Stable Vector Representation of Persistent Homology. [Publisher website]
  in Journal of Machine Learning Research, vol.18(8), pp. 1-35, 2017.
  


• Sofya Chepushtanova, Michael Kirby, Chris Peterson, and Lori Ziegelmeier.
  Persistent Homology on Grassmann Manifolds for Analysis of Hyperspectral Movies.
  [Publisher website] [arXiv]
  in Computational Topology in Image Context (CTIC) 2016, Volume 9667 of Lecture Notes in Computer Science, pp. 228-239; DOI: 10.1007/978-3-319-39441-1_21, online ISBN: 978-3-319-39441-1.


• Sofya Chepushtanova, Michael Kirby, Chris Peterson, and Lori Ziegelmeier.
  An Application of Persistent Homology on Grassmann Manifolds for the Detection of Signals in Hyperspectral Imagery. [Publisher website]
  in Geoscience and Remote Sensing Symposium (IGARSS), 2015 IEEE International, pp. 449-452, 26-31 July 2015, DOI:10.1109/IGARSS.2015.7325797


• Sofya Chepushtanova and Michael Kirby.
  Classification of Hyperspectral Imagery on Embedded Grassmannians. [arXiv]
  in Proc. 6th IEEE Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), Switzerland, 2014.


• Sofya Chepushtanova, Christopher Gittins, and Michael Kirby.
  Band Selection in Hyperspectral Imagery Using Sparse Support Vector Machines. [Publisher website]
  in Proc. SPIE 9088, Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery XX, 90881F (June 13, 2014); DOI: 10.1117/12.2063812.


• Kun Wang, Vineet Bhandari, Sofya Chepushtanova, Greg Huber, Stephen O’Hara, Corey S.O’Hern, Mark D. Shattuck, and Michael Kirby.
  Which Biomarkers Reveal Neonatal Sepsis? [Publisher website]
  in PLoS ONE 8(12), DOI: 10.1371/journal.pone.0082700, December 2013.


• Sofya Chepushtanova and Igor L. Kliakhandler.
  Slow rupture of viscous films between parallel needles. [Publisher website]
  in Journal of Fluid Mechanics, Volume: 573, pp. 297-310, 2007.

PhD Dissertation:

• Algorithms for Feature Selection and Pattern Recognition on Grassmann Manifolds. [CSU website]

Selected Presentations:

• 2020 Virtual Biennial Conference (VBC), June 2020
  Pre-conference workshop Learning from Data - with Anthony Kapolka
  [video lectures and practice sessions]


• TDA: Theory and Applications Conference at Macalester College, St. Paul, MN, June 2017
  Persistent Homology on Grassmann manifolds for Analysis of Hyperspectral Movies [poster]


• Mathematics Department Seminar, University of Scranton, Scranton, PA, March 2016
  Persistent Homology and Its Alternative Vector Representation [invited talk]


• 2015 Luzerne and Lackawanna Counties Mathematics Symposium, Dallas, PA
  Persistent Homology on Grassmann manifolds for Analysis of Hyperspectral Movies [talk]


• 2015 Joint Mathematics Meetings, San Antonio, TX, January 2015
  Sparse Grassmannian Embeddings for Hyperspectral Image Classification [talk abstract]


• SPIE DSS 2014, Baltimore, MD, May 2014
  Band Selection in Hyperspectral Imagery Using Sparse Support Vector Machines [poster]


• Algorithms for Threat Detection Program Review, Boulder, CO, March 2014
  Exploring Uses of Persistent Homology for Hyperspectral Remote Sensing [talk]


• Conference on Data Analysis (CoDA) 2014, Santa Fe, NM
  An Application of Persistent Homology on Grassmann Manifolds to the Detection of Signals in Hyperspectral Imagery [poster]


• Argonne National Laboratory, February 2014
  Data Analysis Methods and Applications: Hyperspectral Band Selection and Data Classification on Embedded Grassmannians [talk]


• Topological Data Analysis Workshop, SAMSI, February 2014
  Set-to-Set Pattern Recognition on Grassmann Manifolds [poster]


• 2014 Joint Mathematics Meetings, Baltimore, MD
  Pattern Classification by Ellipsoidal Machines Using Semidefinite Programming [talk abstract]


• 2013 Front Range Applied Mathematics (FRAM) Student Conference, Denver, CO
  Comprehensive Analysis of Hyperspectral Data using Band Selection based on Sparse Support Vector Machines [talk]


• 2013 Joint Mathematics Meetings, San Diego, CA
  Hyperspectral Band Selection Using Sparse Support Vector Machines [talk abstract]


• 2012 SIAM Annual Meeting, Minneapolis, MN
  Sparse Support Vector Machines for Classification on Grassmannians [talk abstract]


• Conference on Data Analysis (CoDA) 2012, Santa Fe, NM
  Algorithms and Applications of Sparse Support Vector Machines [CoDA website]
  Los Alamos Statistical Sciences Conference Grant winner


• Greenslopes Graduate Student Seminar at CSU, Fort Collins, CO
  Introduction to Support Vector Machines [talk]


• 2005 58th Annual Meeting of the Division of Fluid Dynamics, Chicago, IL
  Theory and Experiments of Slow Rupture of Viscous Films [talk abstract]

Student Poster Presentations

(asterisk * denotes student presenter)

• 2019 Wilkes University Scholarship Symposium, Wilkes-Barre, PA, April 2019
  The Use of Persistence Images as a Topological Order Parameter for Protein Dynamics
  with Brandon Brea* , Daniel Sales*, and Del Lucent


• 2018 Wilkes University Scholarship Symposium, Wilkes-Barre, PA, April 2018
  Topological analysis of protein dynamics using persistent homology
  with Daniel Sales*, Michael O’Brien*, and Del Lucent


• 72nd Annual Eastern Colleges Science Conference (ECSC), Ithaca, NY, April 2018
  Image classification dataset and API design for SVM experimentation
  with Justin Bodner*, Simon Chu*, and Anthony Kapolka


• The 62nd Annual Meeting of Biophysical Society, San Francisco, CA, February 2018
  Topological data analysis of protein dynamics using persistent homology
  with Daniel Sales*, Michael O’Brien*, and Del Lucent


• TecBridge Innovation Conference, Scranton, PA, August 2017
  Topological data analysis of protein dynamics using persistent homology
  with Daniel Sales*, Michael O’Brien*, and Del Lucent


• 71st Annual Eastern Colleges Science Conference (ECSC), Wilkes-Barre, PA, April 2017
  Survey of Image Classification Methods Using an Animal Image Dataset
  with Corey Smithmyer*, Mark Roche*, Abigail Sanders*, and Anthony Kapolka

Service to Profession:

• Program Committee member for the ATDA2019 Workshop: Applications of Topological Data Analysis (Germany), 2019


• National Science Foundation grant proposal review panelist (Alexandria, VA), 2018


• Referee for the IEEE Journals: Geoscience and Remote Sensing Letters, Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Transactions on Knowledge and Data Engineering, and Transactions on Geoscience and Remote Sensing