Research Overview
My primary interest is the Design and Analysis of Algorithms and Computer Science Education. I am particularly interested in Machine Learning algorithms used for pattern matching and prediction.
In the very recent past, my research included applying machine learning algorithms to problems in Bioinformatics, specifically, sequence alignment of DNA and secondary structure prediction of protein sequences and clustering algorithms that facilitate the structural analysis of protein sequences to help determine relationships between the primary structure and the 3D structure.
Currently, I am most interested in Web Design, Multimedia Development and Management Information Systems, which are all new areas for me. I am in the midst of a project to evaluate the effectiveness of video instructions vs written instructions in laboratory-based courses. I have been the webmaster for the Computer Science Department since 2002 and Siena's School of Science since 2007.
Research Links
Bioinformatics
- SCOP: Structural Classification of Proteins
- PDB: Protein Databank
- NCBI: National Center for Biotechnology Information
- Protein Secondary Structure Prediction
Web Design
Protein Structure
A representation of the 3D structure of myoglobin, showing coloured alpha helices. This protein was the first to have its structure solved by X-ray crystallography by Max Perutz and Sir John Cowdery Kendrew in 1958, which led to their receiving a Nobel Prize in Chemistry.
Collaborations & Projects
| Bioinformatics Group, CS Dept., Rensselaer Polytechnic Institute | 2002-present |
| Currently collaborating with Mark Golderg and Malik Magdon-Ismail in the design of machine learning algorithms for secondary structure prediction and classification. Previous research includes the design of maching learning algorithms to enhance DNA sequence comparison (local and global sequence alignment). | |
| Center for Pervasive Computing and Networking, RPI | 2003-2004 |
| Collaborated with Boleslaw Szymanski in the design of a bioinformatics-based algorithm for detecting a masquerade attack, a computer security attack where an intruder assumes the identity of a legitimate user. | |
| Applied Learning Sciences, T.J. Watson Research Center | 1999 |
| Assisted Robert Farrell and Peter Fairweather in the design and implementation of a tutoring adjunct, a computer system that analyzes an operating systems event stream to recognize user tasks and deliver specific help and feedback. | |
Ph.D. Thesis
| A Machine Learning Approach for Designing Dynamic Programmming Algorithms, RPI 2002 | |
|
Dynamic programming is a fundamental algorithmic strategy used to solve many computationally intense
problems. Some of the most recent and notable problems are sequence comparison and alignment, which
are critical tools in the analysis of genetic data. This thesis introduces a new approach for
improving the efficiency of dynamic programming through the use of machine learning. Full Paper [postscript] |
|
Journal Publications
| Discovering Optimization Algorithms Through Automated Learning, Eric A. Breimer, Mark K. Goldberg, David Hollinger, Darren T. Lim, DIMACS Series in Discrete Mathematics and Theoretical Computer Science Vol. 69, pp. 7-27, 2005. | [pdf] |
| A Learning Algorithm for the Longest Common Subsequence Problem, Eric Breimer, Mark Goldberg, Darren Lim, ACM Journal of Experimental Algorithms (online) Vol. 8, Article 4, 2003. | [pdf]
[ps] Source Code [tar] |
| On the Height of a Random Set of Points in a d-dimensional Unit Cube, Eric A. Breimer, Mark K. Goldberg, Brian Kolstad, Malik Magdon-Ismail, Journal of Experimental MathematicsVol. 10, No. 4, pp. 583-597, 2001. | [pdf]
[ps] Source Code [tar] |
Refereed Conference Proceedings
| Intrusion Detection: A Bioinformatics Approach,
Scott Coull, Joel Branch, Boleslaw Szymanski, Eric Breimer,
19th Annual Computer Security Applications Conference,
December 8-12, 2003, Las Vegas Proceedings, pp. 24-33. |
[pdf] |
| A Supervised Learning Approach for Detecting Significant Local Alignments, Eric A. Breimer, Mark K. Goldberg, International Conference on Research in Computational Molecular Biology, April 18-21, 2002, Washington DC, Currents in Computational Molecular Biology, pp. 26 - 27. | [pdf] [ps] |
| Learning Significant Alignments: An Alternative to Normalized Local Alignment, Eric A. Breimer, Mark K. Goldberg, 13th International Symposium on Methodologies for Intelligent Systems, June 27-29, 2002, Lyon, France. Lecture Notes in Computer Science, Vol. 2366, pp. 37-45. | [pdf] [ps] |
| Experimental Evaluation of the Height of a Random Set of Points in a d-dimensional Cube, Eric A. Breimer, Mark K. Goldberg, Brian Kolstad, Malik Magdon-Ismail, Workshop on Algorithm Engineering and Experiments, January 5-7, 2001, Washington DC. | [pdf] [ps] |
| A Learning Algorithm for the Longest Common Subsequence Problem, Eric A. Breimer, Mark K. Goldberg, Darren T. Lim, Workshop on Algorithm Engineering and Experiments, January 7-8, 2000, San Francisco. | [pdf] [ps] |
| A Task-based Architecture for Application-aware Adjuncts, Robert Farrell, Peter Fairweather, Eric Breimer, International Conference on Intelligent User Interfaces, January 9-12, 2000, New Orleans. | [pdf] [doc] |
Technical Reports
| Case Study of a Learning Algorithm for the Longest Common Subsequence Problem, Eric A. Breimer, Mark K. Goldberg, Computer Science Technical Report, Rensselaer Polytechnic Institute, 1998. | [pdf] [ps] |
| Exploring Collaborative Learning in Rensselaer's Classroom-in-the-Round, Robert F.Dugan, Eric A. Breimer, Darren T. Lim, Ephraim P. Glinert, Mark K. Goldberg, Computer Science Technical Report, Rensselaer Polytechnic Institute, 1998. | [pdf] [ps] |