Home Publications Personal Contact

Research

Publications

  • Analytics for Power Grid Distribution Reliability in New York City.
    Cynthia Rudin, Seyda Ertekin, Rebecca Passonneau, Axinia Radeva, Ashish Tomar, Boyi Xie, Stanley Lewis, Mark Riddle, Debbie Pangsrivinij, John Shipman, Tyler McCormick.
    Submitted to the INFORMS Interfaces.
    * Invited paper as the winner of the INFORMS Innovative Applications in Analytics Award.

  • Reactive Point Processes: A New Approach to Predicting Power Failures in Underground Electrical Systems.
    Seyda Ertekin, Cynthia Rudin, Tyler McCormick.
    Submitted to the Annals of Applied Statistics.
    The supplementary file is here:


  • Predicting Power Failures with Reactive Point Processes.
    Seyda Ertekin, Cynthia Rudin, Tyler McCormick.
    In Proceedings of AAAI Conference on Artificial Intelligence (AAAI-2013), Bellevue WA, July 2013

  • Adaptive Oversampling with Active Learning in Imbalanced Data Classification.
    Seyda Ertekin.
    In Proceedings of 28th International Symposium on Computer and Information Sciences (ISCIS), Paris France, October 2013.

  • Active Learning for Imbalanced Learning.
    Josh Attenberg, Seyda Ertekin.
    Book Chapter. Book Title: Imbalanced Learning: Foundations, Algorithms, and Applications. Publisher: Wiley-IEEE. Editors: Haibo He, Yunqian Ma. 2013.

  • Selective sampling of labelers for approximating the crowd.
    Seyda Ertekin, Haym Hirsh, and Cynthia Rudin.
    In Proceedings of AAAI Fall Symposium on Machine Aggregation of Human Judgment (AAAI), Arlington VA, November 2012.

  • Approximating the Crowd
    Seyda Ertekin, Cynthia Rudin, Haym Hirsh.
    Under review at DMKD. Also available as an MIT Tech Report, April 2012.
    Citable URL:http://hdl.handle.net/1721.1/69977

  • Learning to Predict the Wisdom of Crowds.
    Seyda Ertekin, Haym Hirsh, Cynthia Rudin
    In the proceedings of Collective Intelligence 2012, Cambridge MA, April 2012.

  • On Equivalence Relationships Between Classification and Ranking Algorithms.
    Seyda Ertekin, Cynthia Rudin
    Journal of Machine Learning Research (JMLR), October 2011.

  • Approximating the Wisdom of Crowds.
    Seyda Ertekin, Haym Hirsh, Cynthia Rudin
    NIPS Workshop on Computational Social Science and the Wisdom of Crowds, Sierra Nevada, Spain, December 2011.

  • Active Learning for Imbalanced Learning
    with Josh Attenberg and Foster Provost
    To appear as a book chapter in Imbalanced Learning: Foundations, Algorithms and Applications. Editors: Haibo He and Yunqian Ma

  • Non-convex Online Support Vector Machines
    Seyda Ertekin, Leon Bottou, C. Lee Giles
    IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), February 2011.

  • Algorithms for Efficient Learning Systems: Online and Active Learning Approaches.
    Seyda Ertekin.
    Book, VDM Verlag, 2009.
    *Available at Amazon.com.

  • Learning in Extreme Conditions: Online and Active Learning with Massive, Imbalanced and Noisy Data.
    Seyda Ertekin.
    Ph.D. Thesis. Pennsylvania State University, University Park PA, 2009.

  • Adaptive Resampling with Active Learning
    Seyda Ertekin, Jian Huang, C. Lee Giles
    Technical Report, Pennsylvania State University, University Park PA, 2009.

  • Finding Topic Trends in Digital Libraries
    Levent Bolelli, Seyda Ertekin, Ding Zhou, C. Lee Giles
    In Proc. of 8th Joint Conference on Digital Libraries (JCDL 2009), Austin, Texas, June 2009.

  • Topic and Trend Detection in Text Collections using Latent Dirichlet Allocation
    Levent Bolelli, Seyda Ertekin, C. Lee Giles
    In Proc. of 31st European Conference on Information Retrieval (ECIR 2009), Toulouse, France, April 2009.
    *One of the two winners of Google travel award for female computer scientists.
    Published in Lecture Notes in Computer Science, volume 5478/2009.

  • What is trendy? Generative Models for Topic Detection in Scientific Literature
    Levent Bolelli, Seyda Ertekin, C. Lee Giles
    Technical Report, The Pennsylvania State University, University Park, November 2008.

  • Learning on the Border: Active Learning in Imbalanced Data Classification
    Seyda Ertekin, Jian Huang, Léon Bottou, C. Lee Giles
    In Proc. of ACM 16th Conference on Information and Knowledge Management (CIKM 2007), Lisboa, Portugal, November 2007.
    *Also an NEC Laboratories Technical Report, May 2007.

  • K-SVMeans: A Hybrid Clustering Algorithm for Multi-Type Interrelated Datasets
    Levent Bolelli, Seyda Ertekin, Ding Zhou, C. Lee Giles
    In Proc. of IEEE/WIC/ACM International Conference on Web Intelligence (WI 2007), San Jose, California, November 2007.

  • Active Learning for Class Imbalance Problem
    Seyda Ertekin, Jian Huang, C. Lee Giles
    In Proc. of the 30th International Conference on Research and Development in Information Retrieval (ACM SIGIR 2007), Amsterdam, Netherlands, July 2007, (short paper).
    *SIGIR Travel Award Winner

  • A Clustering Method for Web Data with Multi-Type Interrelated Components
    Levent Bolelli, Seyda Ertekin, Ding Zhou, C. Lee Giles
    In proc. of 16th International World Wide Web Conference (WWW 2007), Banff, Alberta/Canada, May 2007, published by ACM, (short paper).

  • Efficient Multiclass Boosting Classification with Active Learning
    Jian Huang, Seyda Ertekin, Yang Song, Hongyuan Zha, C. Lee Giles
    In Proc. of 2007 SIAM 7th International Conference on Data Mining (SDM 2007), Minneapolis, Minnesota, April 2007.
    *IBM Research Travel Award Winner

  • Efficient Name Disambiguation for Large Scale Datasets
    Jian Huang, Seyda Ertekin, C. Lee Giles
    In proc. of 10th European Conference on Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD 2006), pp. 536-544, Berlin, Germany, September 2006.
    Published in Lecture Notes in Computer Science, volume 4213/2006.
    *The poster of this work also won the Best Poster Award in Greater NY Area DB/IR Day which is held in New York University (NYU), New York, October 2006.
    *The media coverage of this paper can be found under the title: The New System Solves The "Who is J. Smith?" Puzzle.


  • Document Clustering Using Sparse Citation Graph Analysis
    Levent Bolelli, Seyda Ertekin, C. Lee Giles
    In proc. of 10th European Conference on Principles and Practice of Knowledge Discovery in Databases (ECML/PKDD 2006), pp. 30-41, Berlin, Germany, September 2006.
    Published in Lecture Notes in Computer Science, volume 4213/2006.


  • Fast Author Name Disambiguation in Citeseer
    Jian Huang, Seyda Ertekin, C. Lee Giles
    IST Technical Report No. 0019. The Pennsylvania State University, University Park, September 2006.
    Longer version of ECML/PKDD 2006 paper.
    *The media coverage of this paper can be found under the title: The New System Solves The "Who is J. Smith?" Puzzle.


  • Fast Classification with Support Vector Machines
    Seyda Ertekin, Léon Bottou, C. Lee Giles
    Grace Hopper Celebration of Women in Computing, San Diego, October 2006.
    Extended abstract of the poster has been selected for ACM Student Research Competition (ACM-SRC) in the same conference.

  • Efficient Support Vector Learning for Large Scale Datasets
    Seyda Ertekin
    In proc. of Doctoral Consortium at Grace Hopper Celebration of Women in Computing, San Diego, October 2006.

  • Active Learning, Loss Function Convexity, and Support Vectors
    Leon Bottou, Ronan Collobert, Seyda Ertekin, Jason Weston.
    NIPS Workshop on Value of Information in Inference, Learning and Decision-Making, Vancouver, Canada, December 2005.

  • Fast Kernel Classifiers with Online and Active Learning
    Antoine Bordes, Seyda Ertekin, Jason Weston, Léon Bottou
    Journal of Machine Learning Research (JMLR), vol. 6, pp. 1579-1619, 2005.
    The source code of LASVM (an online SVM algorithm) is provided at http://leon.bottou.com/projects/lasvm


  • Can Computers Learn Faster?
    Seyda Ertekin
    In proc. of Second COE Research Symposium, The Pennsylvania State University, University Park, April 2005.

  • Comparative Study of Representation of Web Pages in Automatic Text Categorization
    Seyda Ertekin, C. Lee Giles
    IIS-IST Technical Report-62003, The Pennsylvania State University, University Park, June 2003.

  • The Shape of the Web and its Implications for Searching the Web
    Kemal Efe, Vijay Raghavan, C. Henry Chu, Adrienne L. Broadwater, Levent Bolelli, Seyda Ertekin
    In proc. of International Conference of Advances in Infrastructure for Electronic Business, Science, and Education on the Internet, Italy, July 2000.
  • Refereed Poster Presentations

  • Learning to predict the wisdom of crowds.
    Seyda Ertekin, Haym Hirsh, Cynthia Rudin.
    In 19th IMS/ASA Spring Research Conference on Statistics in Industry and Technology, Harvard University, Cambridge MA, June 2012.

  • Cost Effective Crowdsourcing.
    Seyda Ertekin, Haym Hirsh, Cynthia Rudin.
    In New England Machine Learning Day (NEML), Microsoft Research, Cambridge MA, May 2012.

  • Better Algorithms for Energy Grid Maintenance Prioritization
    Seyda Ertekin, Cynthia Rudin
    MITEI Energy Symposium, Massachusetts Institute of Technology (MIT), October 2010.
  • Efficient Name Disambiguation for Large Scale Datasets
    Jian Huang, Seyda Ertekin, C. Lee Giles
    Greater NY Area DB/IR Day, New York University (NYU), New York, October 2006.
    Received the Best Poster Award.

  • Fast Online Classification with SVMs
    Seyda Ertekin, Léon Bottou, C. Lee Giles
    Workshop for Women in Machine Learning, San Diego, October 2006.
    Presented with a spotlight talk in the workshop as well.

  • Fast Classification with Online Support Vector Machines
    Seyda Ertekin, Léon Bottou, C. Lee Giles
    The National Conference of Artificial Intelligence (AAAI) Doctoral Consortium, Boston, July 2006.

  • Scalable Online Support Vector Machines
    Seyda Ertekin, Léon Bottou, C. Lee Giles
    North East Student Colloquium on Artificial Intelligence, (NESCAI), Cornell University, Ithaca, NY, April 2006.


  • Towards Fast Machine Learning Algorithms for Automatic Classification
    Seyda Ertekin, Léon Bottou, C. Lee Giles
    In proc. of Turkish-American Scientists and Scholars Association Annual Meeting, Philadelphia, March 2006.
    Received Young Scientist Grant award from TASSA.


  • The electronic versions of the published papers are made available to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and conditions invoked by each author's copyright.