DIMACS Working Group on Monitoring Message Streams

Retrospective and Prospective Event Detection

This working group meets every two weeks and comprises DIMACS researchers working on the project "Monitoring Message Streams: Retrospective and Prospective Event Detection." The National Science Foundation supports this project through its KDD program. Fred Roberts is the principal investigator. Paul Kantor is the project coordinator. The group is interested in topics such as:

• large-scale text categorization
• streaming data
• dimensionality reduction

Viewing documents marked with a requires a password. Please contact madigan@stat.rutgers.edu, froberts@dimacs.rutgers.edu, or kantorp@cs.rutgers.edu for access.

BBR/BMR/BXR Software

Various versions of the software are at bayesianregression.org

Schedule, Personnel, and Internal Documents

• Spring 2005 Seminar Series
• Fall 2004 Seminar Series
• Spring 2004 Seminar Series
• Fall 2003 Seminar Series
• Members of the Working Group
• Spring 2003 schedule
• Summer 2002 schedule
• Fall 2002 schedule
• Internal documents (requires a separate password)

Papers

• Algorithms for Sparse Linear Classifiers in the Massive Data Setting.
  Balakrishnan, S. and Madigan, D., Journal of Machine Learning Research., 2008, to appear.
• LAPS: Lasso with Partition Search.
  Balakrishnan, S. and Madigan, D., ICDM 2007.
• A Bayesian feature selection score based on Naive Bayes models.
  Eyheramendy, D. and Madigan, D., In: Computational Methods of Feature Selection, H. Liu and H. Motoda, Editors, 2007.
• A Flexible Bayesian Generalized Linear Model for Dichotomous Response Data with an Application to Text Categorization.
  Eyheramendy, S. and Madigan, D. In: IMS Lecture Notes - Monograph Series, Volume 54, Complex datasets and inverse problems: tomography, networks and beyond. Regina Liu, William Strawderman & Cun-Hui Zhang, Editors, 76-91, 2007.
• Decision Trees for Functional Variables.
  Balakrishnan, S. and Madigan, D., ICDM 2006.
• Constructing Informative Prior Distributions from Domain Knowledge in Text Classification.
  A. Dayanik, D. Lewis, D. Madigan, V. Menkov, A. Genkin. (SIGIR 06.)
• Author Identification on the Large Scale
  D. Madigan, A. Genkin, D. Lewis, S. Argamon, D. Fradkin, and L. Ye. CSNA, 2005.
• Exploration Approaches to Adaptive Filtering
  D. Fradkin and M. Littman, January 2005, DIMACS Technical Report 2005-01
• Methods for Learning Classifier Combinations: No Clear Winner.
  D. Fradkin and P. Kantor. Proceedings of ACM Symposium on Applied Computing Information Access and Retrieval Track, 2005.
• Summarizing and Mining Skewed Data Streams.
  G. Cormode and S. Muthukrishnan. SIAM Conf on Data Mining (SDM), 2005.
• Space Efficient Mining of Multigraph Streams.
  G. Cormode, S. Muthukrishan. ACM Principles of Database Systems (PODS), 2005.
• Bayesian Multinomial Logistic Regression for Author identification.
  D. Madigan, A. Genkin, D. Lewis and D. Fradkin, MaxEnt 2005.
• A One-Pass Sequential Monte Carlo Method for Bayesian Analysis of Massive Datasets
  Suhrid Balakrishnan and David Madigan, 2004.
  Presentation Software
• Holistic UDAFs at streaming speeds.
  T. Johnson, G. Cormode, F. Korn, S. Muthukrishnan, O. Spatscheck, D. Srivastava. SIGMOD 2004.
• Diamond in the Rough: Finding Hierarchical Heavy Hitters in Multi-Dimensional Data.
  G. Cormode, F. Korn, S. Muthukrishnan, D. Srivastava. SIGMOD 2004.
• What is New: Finding Significant Differences in Network Data Streams.
  G. Cormode and S. Muthukrishnan. INFOCOM 2004.
• An Improved Data Stream Summary: The Count-Min Sketch and its Applications.
  G. Cormode and S. Muthukrishnan. LATIN 2004.
• Large-scale Bayesian Logistic Regression for Text Categorization (PS) (PDF).
  A. Genkin, D. Lewis and D. Madigan. Technometrics, 2007.
• A Design Space Approach to Analysis of Information Retrieval Adaptive Filtering Systems.
  D. Fradkin and P. Kantor. Proceedings of the Thirteenth ACM Conference on Information and Knowledge Management, CIKM 2004.
• Distinguishing Mislabeled Data from Correctly Labeled Data in Classifier Design.
  S. Venkataraman, D. Metaxis, D. Fradkin, C. Kulikowski, I. Muchnik, International Conference on Tools with AI (ICTAI), 2004.
• DIMACS at the TREC 2004 Genomics Track.
  A. Dayanik, D. Fradkin, A. Genkin, P. Kantor, D. Lewis, D. Madigan, V. Menkov. TREC 2004.
• Combinatorial PCA and SVM Methods for Feature Selection in Learning Classifications (Applications to Text Categorization).
  A.Anghelescu and I. Muchnik. Proceedings of the International Conference on Integration of Knowledge Intensive Multi-Agent Systems, pg. 491-496. ISBN 0-7803-7958-6, 2003.
• Maintenance of Multidimensional Histograms.
  S. Muthukrishnan and M. Strauss. FST & TCS 2003.
• Finding Hierarchical Heavy Hitters in Data Streams.
  G. Cormode, F. Korn, S. Muthukrishnan and D. Srivastava. VLDB 2003.
• What is Hot and What is Not: Tracking Most Frequent Items Dynamically.
  G. Cormode and S. Muthukrishnan. PODS 2003.
• On the Naive Bayes Model for Text Classification (PS) (PDF).
  S. Eyheramendy, D. Lewis, and D. Madigan In Proceedings of The Ninth International Workshop on Artificial Intelligence and Statistics, C.M. Bishop and B.J. Frey (Editors), 332-339, 2003.
• Experiments with Random Projections for Machine Learning.
  D. Fradkin and D. Madigan. In Proceedings of KDD-03, The Ninth International Conference on Knowledge Discovery and Data Mining, 2003.

Project Reports

• Constructing Informative Prior Distributions from Domain Knowledge in Text Classification
  Aynur Dayanik, David D. Lewis, David Madigan, Vladimir Menkov, and Alexander Genkin, January 2006.
• Simulated Entity Resolution by Diverse Means: DIMACS Work on the KDD Challenge of 2005
  Andrei Anghelescu, Aynur Dayanik, Dmitriy Fradkin, Alex Genkin, Paul Kantor, David Lewis, David Madigan, Ilya Muchnik and Fred Roberts, December 2005, DIMACS Technical Report 2005-42.
• Decision Trees for Functional Variables
  Suhrid Balakrishnan and David Madigan, 2006.
• A Flexible Bayesian Generalized Linear Model for Dichotomous Response Data with an Application to Text Categorization
  Susana Eyheramendy and David Madigan, 2006
• Bayesian Multinomial Logistic Regression for Author Identification
  David Madigan, Alexander Genkin, David D. Lewis, and Dmitriy Fradkin, August 2005.
• DIMACS at the TREC 2005 Genomics Track
  Aynur Dayanik, Alexander Genkin, Paul Kantor, David D. Lewis, and David Madigan, 2005.
• Sparse Logistic Regression for Text Categorization
  Alexander Genkin, David D. Lewis, and David Madigan. April 2005
• Polytomous Logistic Regression
  David D. Lewis, October 2004.
• Project Overview (PS)
  Paul B. Kantor and Fred S. Roberts, April 2003.
• A design space approach to analysis of IR/AF systems
  Dmitriy Fradkin and Paul Kantor, February 2004.
• Methods for learning classifier combinations: No clear winner
  Dmitriy Fradkin, Paul Kantor, and Kwong Bor Ng, February 2004.
• Optimization of SVM in a Space of Two Parameters: Weak Margin and Intercept.
  Andrei V. Anghelescu and Ilya B. Muchnik, May 2003.
• Prospective Data Fusion for Batch Filtering
  Andrei Anghelescu, Endre Boros, Dmitriy Fradkin, David D. Lewis, Vladimir Menkin, David J. Neu, Kwong Bor Ng., and Paul Kantor, May 2003.
• Feature Selection and Batch Filtering
  Endre Boros and David J. Neu, May 2003.
• Sparse Bayesian Classifiers for Text Categorization
  Susana Eyheramendy, Alexander Genkin, Wen-Hua Ju, David D. Lewis, and David Madigan, May 2003.
• A Closer Look at Nearest Neighbor Text Classification.
  David D. Lewis and Vladimir Menkov, May 2003.
• Prospective Data Fusion for Batch Filtering (PS) (PDF)
  Andrei Anghelescu, Dmitiry Fradkin, David D. Lewis, Vladimir Menkov, Kwong Bor Bg, and Paul Kantor, February 2003.
• A Closer Look at Nearest Neighbor Classification (PS)
  David D. Lewis and Vladimir Menkov, January 2003.
• Towards a Formal Framework for Adaptive Filtering (PS) (PDF)
  Paul Kantor and David Madigan, December 2002.
• Towards Fast, Effective Nearest Neighbor Text Classification (PS) (PDF)
  David D. Lewis, S. Muthukrishnan, R. Ostrovsky, M. Strauss, November 2002.
• Distances from Sketches (PS) (PDF)
  Martin Strauss, November 2002.
• Sparse Bayesian Classifiers for Text Categorization - Batch Learning (PS) (PDF)
  Alexander Genkin, David D. Lewis, and David Madigan, November 2002.
• Sparse Bayesian Classifiers for Text Categorization - Online Learning (PS) (PDF)
  Wen-Hua Ju and David Madigan, November 2002.
• Combinatorial Clustering for Textual Data Representation in Machine Learning Models
  Andrei Anghelescu and Ilya Muchnik, November 2002.
• An Experimental Evaluation of a Combinatorial Clustering Model for Textual Data Representation
  Andrei Anghelescu and Ilya Muchnik, November 2002.
• A Simple, Coarse Vector Nearest Neighbors Algorithm (PS) (PDF)
  S. Muthukrishnan, November 2002.
• Some Statistical Analysis Algorithms on Data Streams (PS) (PDF)
  S. Muthukrishnan, November 2002.
• Notes on Term Selection
  Endre Boros, November 2002.
• Training a Rocchio classifier for adaptive filtering (PS) (PDF)
  Andrei Anghelescu, David D. Lewis, Vladimir Menkov, and Paul Kantor, TREC 2002.
• Feature Selection and Batch Filtering (PS) (PDF)
  Endre Boros and David J. Neu, TREC 2002.
• CCR-DIMACS Workshop
  Bob Grossman, Paul Kantor, and Muthu Muthukrishnan, June 2002.

Online Tools

• Current Comparison Tool
• Eye of Deity Tool
• Histogram/ROC Tool

Project Deliverables

• November 2005 Deliverables
• April 2005 Deliverables
• February 2004 Deliverables

Project Presentations

• August 2006 Site Visit
• September 2004 PI's Meeting Overview Presentation
• April 2005 Site Visit
• July 2004 Site Visit
• February 2004 Site Visit
• a href="MMSProject11-18-03rev3.ppt">Monitoring Message Streams: Algorithmic Methods for Automatic Processing of Messages (PPT)
  Fred Roberts, November 2003.
• Monitoring Message Streams: Retrospective and Prospective Event Detection (PPT)
  Fred Roberts, December 2002.
• Monitoring Message Streams: Future Work (PPT)
  Fred Roberts, December 2002.
• Combinatorial Clustering for Textual Data Representation (PDF)
  Andrei Anghelescu and Ilya Muchnik, December 2002.
• Towards Fast Algorithms for Nearest-Neighbor Searching (PPT)
  Rafail Ostrovsky, December 2002.
• Nearest Neighbor Text Classification (PPT)
  David D. Lewis, S. Muthukrishnan, R. Ostrovsky, and M. Strauss, December 2002.
• Infrastructure: Software and Data (PPT)
  Andrei Anghelescu, Paul Kantor, David D. Lewis, and Vladimir Menkov, December 2002.
• Fusion of Methods (PPT)
  Paul Kantor and Dmitriy Fradkin, December 2002.
• How do we measure success? (PPT)
  Paul Kantor, December 2002.
• Towards a Formal Framework for Adaptive Filtering (PPT)
  Paul Kantor and David Madigan, December 2002.
• Bayesian Methods for Text Categorization (PPT)
  Susana Eyheramendy, Alex Genkin, Weh-Hua Ju, David D. Lewis, and David Madigan, November 2002.
• Streaming Data Analysis (PPT)
  S. Muthu Muthukrishnan, November 2002.
• Randomized Projection Methods (PS)
  Martin Strauss, November 2002.
• Feature Selection
  Endre Boros, November 2002.
• Project presentation (PPT), August 2002
• Project presentation (PPT), May 2002
• Project proposal (Word), March 2002

Informal Results Pages

• Fusion
• Bayes

Related Events

• TREC Conferences
• June 17-22, 2002 CCR/DIMACS Workshop/Tutorial on Mining Massive Data Sets and Streams: Mathematical Methods and Algorithms for Homeland Defense

Other pointers to relevant papers by group members:

• Madigan
• Muthukrishnan
• Ostrovsky 1 2 3 4