Dennis F. Kibler



Professor Emeritus


Artificial Intelligence


Bren Hall 4044


Office Hours:

By appointment

Genomic Analysis via Machine Learning Methods

These research projects are being done in collaboration with a number of faculty from the School of Biological Science and the College of Medicine. For validation of the computational approach each project is focussed on a particular genome.

  • Promoter modelling. Collaborator: Ming Tan. Students: Hilda Yu and Johnny Ackers. We have developed a bimotif variable gap model that has been successful at predicting the sigma-28 promoter site in Chlamydia and E. coli Extensions of this model will be applied to search for dimer binding sites.
  • Hydra database Collaborators: Hans Bode, Robert Steele, and Steve Hampson. Analysis and data from Hydra ests, as they are being sequenced. Currently we have about 140,000 ests available and new sequences are added monthly. The main goals are to identify genes, their functions, and their evolutionary histories.
  • Discovery of Chlamydial Transcription factor Binding Sites Collaborator: Ming Tan. Student: Bob Chan. Chlamydia is an unusual bacterial that lives entirely within human cells. We are developing algorithms that combine various forms of evidence to identify the transcription binding sites that control the early and late genes.

In general our goal is to develop programs that can use the available data to help determine biological significant substructures or patterns in genomes.

Current Graduate Advisees and Projects

        Johnny Akers (Bio phd student): Advisor Ming Tan. Identifying regulatory elements in Chlamydia arising from dimerization.

        Martin Brandon Advisor Pierre Baldi MitoMap

Graduate students completing with doctorate

    • Bob Chan (2007) Discovery of Local Patterns in DNA that Predict CRMs and Protein Structural Similarity
    • Hilda Yu (2006) co-advisor with Ming Tan
    • Catherine Blake (2003), Assistant Professor, University of North Carolina, Chapel Hill
    • Yuh-Jyh Hu (1999), Assistant Professor in Computer Science and Engineering Department, Tatung University, Taipei.
    • Piew Datta (1997), Researcher at GTE Research Laboratory.
    • Pedro Domingos (1997), Associate Professor, University of Washington.
    • Pat Murphy (1996), Reseacher Scientist at JPL.
    • David Ruby (1993), Independent Consultant.
    • David Aha (1990), Researcher at Naval Research Laboratory.
    • Etienne Wenger (1990), Research scientist at the Institute for Research on Learning, Xerox Park
    • Rogers Hall (1990), Associate Professor at Vanderbilt, Department of Education.
    • Douglas Fisher (1987), Associate Professor at Vanderbilt University.
    • Paul Morris (1984), Researcher at Intellicorp, Menlo Park, California.
    • Bruce W. Porter (1984), Professor at University of Texas.
    • Steven Hampson (1983), Research Scientist at UCI.
    • John Conery (1983), Professor at University of Oregon.
    • Stephen Fickas (1983), Professor at University of Oregon.

Notable Undergraduates

    • Kimberly Ferguson(2003) Computing Research Award (honorable mention)

Undergraduate Advisees and Projects

    • Frank Chen: Building a monad and dyad motif detector.
    • Timothy Uy: (High school student) Evaluating Genetic Algorithm on BSAT
    • Bach Ho: Extending Weka with improved Nearest Neighbor and k-means algorithms
    • Gary Suh: Diagnose Colon cancer using Protein expression data (Ciphergen)
    • Vishahk: Creating a web-interface and mysql database for HYDRA data.
    • David Hu: Building a dimer binding site detector for chlamydia trachomatis.

Course Offerings:

Cosmos: Data Mining

    • Course for Gifted High School Students
    • Geometric and intuitive approach to three problems in Data Mining
      • Classification
      • Regression
      • Clustering
    • Course will use the freely available Weka software
    • Prerequisites: Comfortable with algebra, geometry and computers
    • No programming experience required

Cosmos (2005) BioInformatics: Understanding our Genome

The goal of this course is to introduce students to the bioinformatics research paradigm by studying three significant problems in understanding the genome. The paradigm consists of starting with a biological problem, biological data and biological knowledge and then building an approximate computational model. The model is then applied to real data and its predictions are then validated by additional biological experiments. Ideally this cycle repeats. The course will begin with the problem of identifying genes from sequence data using the GENSCAN program. After genes are identified, their function will be hypothesized using the BLAST program. How genes are regulated will be answered using programs for finding surprising subsequences. In all of these studies students will work with real data using programs freely available on the web. The ideas underlying the algorithms, their limitations, and their connection to biology will be stressed.

Student Projects with draft abstracts

    • Homosexuality: Determined by genetics or social environment by Mansi Shah and Wendy Kim
    • Applying Microarray Technology to predicting Adverse drug reaction in children with acute lymphoblastic leukemia by Ruwani Ekanayake, Amy Henry, and Brittany Horth
      Abstract: Pharmacogenomics, the study of how an individual's genetic inheritance affects the body's response to drugs, has expanded exponentially since the completion of the Human Genome Project in 2003. Now, thanks to emerging microarray technology, pharmacogenomics is being applied with great success to cancer research. Our project will explore the benefits of microarray technology in pharmacogenomics, specifically as applied to cancers.
    • Stem Cells by Micheal Jenkins, Uchechukwu Nnadi, and Tom Garrett
    • Predicting SNPs and Their Effects on Proteins using SIFT and Polyphen by Eve Shih and Kuo
    • Predicting Leukemia Classes Based on gene expression data using WEKA by Alex Doo, Stehpanie Lang . and George Quinonez
    • Similarity Sequencing and its use for Generation of Phylogenetic Trees by Ivan Cvitkovic and Daniel Kaufman
      Abstract: We plan to explore the concept of similarity sequencing. Through our research, we will discuss the different algorithms for similarity sequencing and a basic description of how they work. Additionally we will discuss the application of similarity sequencing for the generation of phylogenetic trees. During our project we will generate two separate phylogenetic trees, based on different proteins.
    • Pharmacogenomics and Bioinformatcis of Long QT Syndrome by Daryl Serrano and Carlos Palacios .
      Abstract: Long QT syndrome is a genetic disease that affects the heart. One's heart has electrical activity that is made by a flow of ions. This causes the heart to beat normally. People with LQTS have a slower QT interval (heartbeat) than normal. People can acquire LQTS through prescription/over the counter medications or inherit the disease from their parents at birth. LQTS can cause many symptoms in a person that can greatly affect their everyday lives. However, there are treatments for LQTS that help people live normal lives. Acquired LQTS is caused by medications that include antihistamines, antidepressants, mental illness medications, heart medications, etc. LQTS is also congenital, or caused by a mutation in the gene that forms ion channels. There are different classifications of LQTS, such as LQT1, LQT2, LQT3, LQT4, and LQT5. These different forms of LQTS affect different ion channels in the human heart. When LQTS slows one's heart beat, one is likely to faint, have a seizure, or die. One will most likely experience these symptoms when exercising or becomes emotionally excited. Although LQTS may be fatal, doctors have found treatments that help shorted the QT interval. and allow patients to live normal lives.
    • How Breast Cancer response to chemotherapy by Amanda Farrar and Rosalinda Ruiz
      Abstract:< Chemotherapy is very important in metastasis Breast Cancer. Doxorubicin, epirubicin, Paclitaxel, and docetaxle are the most active chemical drugs used in Breast Cancer treatment. Activating tumor-suppressor genes are emerging as important targets for therapy. Researchers test various models to see which combinations of drugs improve Breast Cancer treatment. By understanding the molecular processes underlying Breast Cancer, researchers hope to create specific drugs that will prevent the growth of Breast Cancer. The malfunction of genes allows cancer to spread to various parts of the bodies. Understanding these processes will help find a cure for this disease.
    • Title: What is DEB and how is it treated by Maritza Navarro and Alexandria Magallan
      Abstract: Dystrophic epidermolysis bullosa (DEB) is caused by a mutation in the collagen, type VII alpha 1 protein, which is found in chromosome 3. Type VII collagen, is a protein that helps keep your skin intact. Those who have dystrophic epidermolysis bullosa are usually diagnosed at birth. Patients with the disease have "butterfly skin" which is so thin that any minor trauma to the skin causes it to blister and scar. The disease is a rare genetic disease that can be both recessive and dominant. If a person has dominant DEB then he or she has no real cure except for the nurture from those who care for them. But those who have recessive DEB, may have a chance since there has recently been 4 cases of skin graphs, or cutaneous pinch grafts, that helps to heal the wounds of the patients.
    • Similarity Sequencing and its use for phylogenetic trees by Ivan Cvitkovic and Daniel Kaufman.
    • The genetic causes and physiological effect of sickle cell anemia by Alda Caan and Blanca Trujillo
    • How gene therapy may be applied to the curing genetic diseases by Michael Jenkins and Uchechukwu (Uli)Nnadi.

Freshman Seminar: Artificial Intelligence: Is it for Real?

    • Discussion of historical and significant papers in Artificial Intelligence dealing with the creating and evaluation of computational artifacts that do or do not exhibit intelligence. Students will be encouraged to suggest their own approaches to the problems. Brainstorming and constructive evaluation will be encouraged.
    • Prerequisite: ICS21 with a grade of B.
    • Readings: Significant papers will be chosen, corresponding to student interest. If you miss class, then you can pick up the paper outside my office. Also if you miss class, you are required to write a one paragraph comment on the reading which you should send to me via email.
    • First Paper: the Turing test (1950)
    • Syllabus
    • Meetings Wednesdays: 3-3:50, CSE 310.
    • First Class: April 2; Last Class: June 4
    • Grading: Grades are based on class participation and written assignments. Attendance is not sufficient. Each week you are expected to hand in a one paragraph comment on the reading.

H22 Honors Introduction to Computer Science II

    • Introduction to basic abstract data structures and associated algorithms, including their implementation, selection, and complexity. Data structures include lists, stacks, queues, tables, and trees.
    • Prerequisite: H21 or consent of instructor
    • Required Texts:
      • Data Structures and Algorithms with Object-Oriented Design Patterns in Java by Bruno Priess.
      • Core Java (2nd ed) by Horstmann and Cornell
    • Recommended: Any book on Java that works for you. Many students like Core Java 2. Java Texts .
    • Syllabus
    • Ethics

Honors 23 Problem Solving and Data Structures

    • Further analysis of basic and non-basic data structures and associated algorithms. With respect to representation, covers arrays, lists, trees and graphs. With respect to algorithms covers recursion, divide-and-conquer, backtracking, and dynamic programming.
    • Prerequisite: H22 or consent of instructor
    • Required Text: Data Structures and Problem Solving in Java
    • Author: Mark Allen Weiss
    • Recommended for Swing: Up to Speed with Swing by Steven Gutz.
    • Syllabus
    • Homework Details
    • Ethics

ICS 23 Problem Solving and Data Structures

    • Further analysis of basic and non-basic data structures and associated algorithms. With respect to representation, covers arrays, lists, trees and graphs. With respect to algorithms covers recursion, divide-and-conquer, and separate-and-conquer,
    • Prerequisite: ICS22 or consent of instructor
    • Required Text: Data Structures and Problem Solving in Java
    • Author: Mark Allen Weiss
    • Suggested for Gui's: Up to Speed with Swing by Steven Gutz.
    • Syllabus
    • Homework Details
    • Ethics

171 Introduction to Artificial Intelligence (4)

    • The course is divide into four major topics and we will spend about two weeks on each topic. The major topics are: Problem Solving via search, Logical reasoning in propositional and first-order logic, Probabilistic Reasoning and Learning.
    • Prerequisites: ICS 52 and and Mathematics 2A-B and 67.
    • Text: Artificial Intelligence: A Modern Approach
    • Authors: Stuart Russell and Peter Norvig
    • Grading. There will be 4 quizzes,2 coding assignments, and 3 written homeworks. The quizzes will each count 15% of your grade. The lowest homework/coding score will be dropped. Scores on late homework will be reduced by 20% per day.
    • Class Lectures: MWF 2:00-2:50 CS174
    • Time: June 28 - Sept 3.
    • Teaching Assistant: Rajyashree Mukherjee
    • Course email:
    • Discussion Section: Wed 3:00-3:50 CS174
    • Syllabus and Lectures
    • Ethics
    • Homework Details Future homeworks will be collected in class and returned via the distribution center.

172 Programming Techniques in Artificial Intelligence (4) W.

    • The study of the object-oriented design as applied to AI algorithms and representations. The goal is to create an in-depth understanding of some of the important AI approaches by coding various algorithms in an object-oriented language. On the coding side, we will examine graphical displays, user-interfaces, and code libraries. On the AI side, we will implement algorithms for problem-solving, optimization, decision-making, and learning. There will be three to five coding/design assigments. The code will be in Java or Python.
    • Prerequisites: ICS 171, knowledge of object-oriented programming
    • Texts: Any Java text you like plus any AI text you like.
    • Syllabus plus fuller description of course.
    • Programming Language: Java
    • Ethics

CS 174 BioInformatics (4) S.

    • Meetings: M-W-F: 11-11:50 ELH 110
    • Office Hours: M-W: 9-10 and by appointment. Room 414D CS
    • First Class: Monday, April 3
    • Last Class: Friday, June 9
    • Final: Friday, June 16, 8am -10am
    • Questions: Do not hesitate to ask questions in class, in my office hours or by email. Do hesitate to ask questions on the weekends - that's family time for me. Untimely questions may not be answered.
    • Text: Fundamental Concepts of BioInformatics by Dan Krane and Michael Raymer
    • Course Mailing List: TBD
    • Teaching Assistant: Daniel Sanchez (
    • Teaching Assistant Office hours: Wed- Fri, 10-11, TA office in the ICS Trailers
    • Grading: There will be eight assignments plus one quiz and a final. The lowest score on a assignment will be dropped. The final will count 20% of your grade, the quiz 10% and the best seven of your assignment scores will each count 10%. Homeworks can be turned in one day late for 1/2 credit. All homeworks are due a week after the assignment on Monday by 10 am. If it is turned in at 10:01, then you can only get at most 1/2 credit. Homeworks will be deposited using the checkmate program.
    • Course Goals: Bioinformatics is the study of biological problems via computational tools. The goal of the course is to provide students with sufficient biological knowledge and computational methods that they can use, understand, and possibly generate algorithms that are appropriate for available biological data. This course will concentrate on approaches that deal with the most voluminous and accurate data, namely DNA data, protein data, and gene-expression data. As part of the course students will use current tools on existing databases to a) locate genes and determine their function, b) build phylogenic trees, c) locate regulatory elements and d) predict protein structure.
    • Overview of Course: Problems in Biology and Computational Approaches
      Week 1: Basic structure of DNA, RNA, genes and proteins.
      Week 2&3: Methods: Dot Matrices, Local, Global and Multiple Alignment algorithms.
      Problems: Gene identification and gene function
      Week 4&5. Methods: Tree building: UPGMA, clustering, maximum likelihood
      Problems: Tree of Life: fitting all organism into an evolutionary history
      Week 6&7: Methods: Pattern discover by search (exhaustive and heuristic)
      Problems: Gene regulation in Prokaryotes and Eukaryotes
      Week 8&9: Methods: Dynamic Programming, machine learning
      Problems: Determine Secondary and Tertiary structure of RNA and Proteins.
      Week 10: Methods: Machine Learning
      Problem: Determining proteomic disease diagnoses.
    • Course Workload: Expect weekly assignments. You will have four assignments where you implement basic algorithms for a particular biological problem. These assignments will alternate with using existing, more sophisticated algorithms for related tasks. In particular you will write algorithms for finding genes, determining similarity between DNA and protein sequences, building evolutionary trees, and locating regulatory elements. I will provide a sketch or design for these algorithms so that they are all implementable in one afternoon plus, at least for me, another afternoon for debugging. Only the next homework assignment is guarantee to be correct - other assignments may change as the class progresses.
    • Quiz: Multiple choice and fill in the blank. Monday of the second week of classes. This will be based on chapter 1 of the text (pages 1-20) plus the lectures. For computer scientist learning the basic vocabulary of molecular biology is somewhat difficult. I recommend reading the chapter at least twice, making note of the important concepts.
    • First Assignment to hand in: Due the beginning of the 3rd week of classes. The assignment has two parts for submission: a coding part and question part. The assignment is due by 10am on Monday of the third week of classes. In general assignment are due by 10am on Monday of week after the assignment is made. Assignments are to be deposited in the appropriate folder using checkmate. Most of you have already used this software, but just in case: To set up for electronic submission, go to, log in with your UCInet ID, choose "Course Listing" for "Spring 2006," click "Go" next to ICS 174, and then click "List me for this course." For the answers to questions, you will submit a word file hwkn.doc file. For the coding part, when they occur, submit a single To do this you will need collect your java files into a single file. You may also do the homework in Python. I reserve the right to change assignments as the course progresses.
    • Assignment Details

CS 175A Projects in Artificial Intelligence (4) S.

                                 Meetings: MWF 3:00-3:50 SE2 1304

                                 Office Hours: MW before and after class and by appointment.

                                 Course Mailing List:

                                 Course Project web site TBD

                                 Final paper and code due TBD.

                                 Teaching Assistants: James Worcester ( Office hours: TBD

                                 Poster Presentations (Power-point) with demonstration; Date to be decided.

                                 First Class: September 23 Last Class: December 2

                                 Work Load: Several presentations, interim documents, design documents, website, power-point presentation, Java code, and a final paper. The final paper should include at least two references, either to papers or texts. The final presentation will be done in PowerPoint and should include a demonstration and evaluation of the program. Each project will have an associated WebSite

                                 Grading: All students in a group will assign credit to one another. The individual grade will be based on the total project grade (decided by me) distributed as group members decided.

                                 In this course students will build Java programs that demonstrate AI techniques. The potential topics include Expert Systems, Natural Language Processing, Problem Solving, Search, Game Playing, Learning, Reasoning, Perception, etc. A more specific but still incomplete list of potential projects is listed under AI projects. One common approach would be to implement some AI method and apply it to some domain. The AI text by Russell and Norvig provides many examples of possible projects.

Each student will be involved in one project done in a group of three people. In the project each object will be identified with its author. All students are responsible for some of the code. The programs generated will be made public so that all can view, use, and evaluate. Students will be responsible for maintaining their programs. There will be an emphasis on design so every project will go through a design review in which the entire class participates. Each group will be responsible for several presentations: the basic goal, the history on the problem, the design, and the final demonstration with evaluation. Each group is also responsible for a report that corresponds to each presentation.

                                 Prerequisites: ICS 171, knowledge of Java

                                 AI Projects.

                                 Work Schedule.


202 Seminar in Research in ICS (2) F.

Graduate orientation program and colloquium series. Includes talks by ICS faculty in all areas about their current research. Satisfactory/Unsatisfactory Only. Formerly ICS 295.

209 Seminar in Bioinformatics(2)

Graduate seminar in which recent papers and research are discussed and analyzed. Papers will concentrate on analysis of genomic and gene expression data. Supporting papers on biology, machine learning, and statistics may also be included. It is likely that each enrolled person will present two papers. The first presentation will be informal and will likely include necessary background material. The second presentation will be a formal one, done either via PowerPoint or overheads. During the second presentation background material can be assumed.

                                 ICS 249 Mondays12-2pm (Note new place and time)

                                 Bioinformatic Papers

CS 270A: Introduction to Artificial Intelligence

                                 Introduction to basic AI representations and algorithms, problem solving, planning, logical and probabilistic reasoning, natural language processing and learning.

                                 Text: Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig 2nd Edition. I suggest ordering this from Amazon.

                                 Lecture Time: Tues/Thurs 9:30 - 10:50 Room: cs253

                                 Office hours: 11-12 tues/thurs and by appointment.

                                 First class: Sept 30 2003. Last class: Dec. 4 2003.

                                 Grading: Three coding assignments and a final: all weighted equally.

                                 Final: Thursday Dec 11 8am-10am

                                 Click here for lectures and assignments.

CS273: Machine Learning (4).

                                 Computational approaches to learning, concentrating on classification and regression. Covers standard learning representations (rules, decision trees, instances, linear threshold units, neural nets, etc), their representation, limitations, and evaluation.

                                 Prerequisite: ICS 270A. Formerly ICS 275.

                                 Required Text: Introduction to Data Mining

                                 Authors: Pang-Ning Tan, Michael Steinbach, Vipin Kumar

                                 Publisher: Addison-Wesley ISBN 0-321-32136-7

                                 Recommended Text: Data Mining: Practical Machine Learning Tools and Techniques with Java Implementations. 2nd Edition.

                                 Authors: Ian H. Witten and Eibe Frank. Publisher: Morgan Kaufman

                                 Recommended Text: Machine Learning by Tom Mitchell. Publisher: McGraw-Hill

                                 First Class: Jan 10. Last class: Thursday March 16

                                 Final: Thursday March 23 8am - 10am

                                 Lectures: tues-thurs 9:30-10:50

                                 Room: CS-213

                                 Office Hours: tuesday/thursday after class and by appointment.



277A Representations and Algorithms for Molecular Biology(4)

o                                The primary goal of this class is to introduce molecular biologists to computer science and computer scientist to molecular biology. It is not expected that biologist will become programmers, but they will learn what might be accomplished with computational analysis. Nor is it expected that computer scientists will conduct biological experiments, but they will learn enough biology to understand the important problems in biology that are addressable by computational means.

                                 Required Text: Bioinformatics: Sequence and Genome Analysis by David W. Mount 2nd Edition

                                 Recommended Text: Introduction to Computation Molecular Biology by Sebutal and Meidanis

                                 Recommended (undergraduate text): Fundamental Concepts of Bioinformatics by Krane and Kramer

                                 First Class: Thursday Jan 8, 2006. Last class: Thursday March 16

                                 Final: Thurs Dec 12, 1:30-3:30pm

                                 Meetings: Tuesday/Thursday 9:30- 10:50 CS253

                                 Course Work: Weekly readings from the text and papers, a few homeworks, and a final project.

                                 Course Mailing List:

                                 Project: A joint project and presentation teaming a biologist with a computer scientist.


280 Seminar in Computational Biology

This is an advanced course that concentrates on recent computational methods that aid in the functional analysis of genomes. Papers will be drawn primarily from Bioinformatics, Proceedings of the Conference on Intelligent Systems for Molecular Biology, Journal of Molecular Biology, Proceedings of the National Academy of Science, Science, and the Journal of Computational Biology. The course will focus on algorithms that find patterns in DNA, particularly those that relate to gene regulation, such as finding regulatory elements, finding promoters, and clustering gene expression data. There is no text. We will all be reading, discussing, and presenting papers as well as any research-in-progress.

Java Demos:

o                                Traveling Salesman. Not available now.
This Java (JDK1.0) program illustrates the use of hill-climbing to solve the traveling saleman program. The program is simple so that the general ideas can be understood. Many valuable extensions are possible. The program only uses one operator, that of uncrossing edges that intersect. Additional operators are useful. The general problem of defining useful operators for hill-climbing is unsolved. Simulated annealing, multiple restarts, better initialization would all be useful. If your browser executes JDK1.1, then you might prefer the following similar demo: Not available now.

o                                Dynamic Programming. Dynamic Programming
This program illustrate the use of dynamic programming to find the minimum edit distance between two strings. This distance depends on the cost one associates with various edit operations. It can be used for spell checking but major applications are in comparing amino acid and nucleotide sequences. If you look at the source,the main action is in the routine doSimMatch. This was one of my first Java programs, so the program can be greatly improved.

o                                Kmean clustering. Not available now.
You'll see that is doesn't always work. Requires JDK1.1.

o                                Valdimir Vapnik's applet on Support Vector Machines . This is a beautiful applet that displays the probability density function associated with the generated decision surface.

o                                N Queens Problem. Not available now.
N Queens problem is solved by local improvement or repair search. This same method is most applied to scheduling problems, where it can generate anytime and approximate solutions. Coded in ics175 by Son Tran.



In silico prediction and functional validation of sigma-28-regulated genes in Chlamydia and E. coli. Yu, H.H.Y., Ming Tan, and Dennis Kibler. Journal of Bacteriology. Online at JB01082-06


Using Hexamers to Predict Cis-Regulatory Modules in Drosophila. Bob Chan and Dennis Kibler, BMC Bioinformatics. 6: 262, October 2005.


A horizontally transferred protist gene in the Hydra genome Figure 1 Current Biology. Robert E. Steele, Steven E. Hampson, Nicholas A. Stover, Dennis F. Kibler, Hans R. Bode. Volume 14, number 8.


Using DNA MicroArrays to Identify SP1 as a Transcription Regulatory Element of Insulin-Like Growth Factor in Cardiac Muscle Cells. Circulation Research. Tao Li, Yung-Hsiang Chen, Tsun-Jui Lui, Jia Jia, Steven Hampson, Yue-Xin Shan, Dennis Kibler, Ping H. Wang. pp 1-35. 2003

Evaluating Representations for the Shine-Dalgarno Site in Escherichia coli Steven Hampson and Dennis Kibler. TR#03-14. School of Information and Computer Science. University of California, Irvine.


Characterizing the E. coli Shine-Dalgarno Site: Probability Matrices and Weight Matrices Dennis Kibler and Steven Hampson, International Conference on Mathematical and Engineering Techniques in Medicine and Biological Science ( METMBS-2002). pp. 358-364.

Distribution Patterns of over-represented k-mers in non-coding yeast DNA Steven Hampson, Dennis Kibler, and Pierre Baldi, BioInformatics. vol. 18 no.4 pp. 513-528.


Learning Weight Matrices for Identifying Regulatory Elements, Dennis Kibler and Steven Hampson, International Conference on Mathematical and Engineering Techniques in Medicine and Biological Science ( METMBS-2001). pp. 208-214.

Bay, S. D., Kibler, D., Pazzani, M. J., and Smyth, P. (2001). The UCI KDD Archive of Large Data Sets for Data Mining Research and Experimentation. In Information Processing Society of Japan Magazine. Volume 42, Number 5, pages 462-466. English language version reprinted in SIGKDD Explorations. Volume 2, Issue 2, pp. 81-85, 2000.


Analysis of Yeast's ORF Upstream Regions by Parallel Processing, Microarrays, and Computational Methods. Steve Hampson, Pierre Baldi, Dennis Kibler, and Suzanne Sandmeyer. Tenth International Conference on Intelligent Systems for Molecular Biology ( ISMB-2000). pp. 109-201.

Combinatorial Motif Analysis and Hypothesis Generation on a Genomic Scale , with Yuh-Jyh Hu, Suzanne Sandmeyer, Calvin McLaughlin. BioInformatics, Vol 16, 222-232.


Minimum Generalization via Reflection: A Fast Linear Threshold Learner , with Steven Hampson, Machine Learning 37 pp. 51-73. 1999.

Detecting Motifs from Sequences with Yuh-Jyh Hu and Susan Sandmeyer, International Conference on Machine Learning 1999.

Symbolic Nearest Mean Classifiers , with Piew Datta, AAAI-97.

Learning Symbolic Prototypes , with Piew Datta, ICML-97.

GalaII: Integrating Construction of Boolean and Prototypical Features , with Yuh-Jyh Hu, ECML-97 (in press).

A Generative Approach to Constructive Induction , with Yuh-Jyh Hu, AAAI-96.

``Plateaus and Plateau Search in Boolean Satisfiability Problems: When to Give Up Searching and Start Again,'' with Steven Hampson. DIMACS Challenge, 1995.

Learning Prototypical Concept Descriptions , with Piew Datta, Twelfth International Conference on Machine Learning.

``Learning recurring subplans'' with David Ruby. in Machine Learning Methods for Planning, Minton, S., 466--497, Morgan Kaufman, 1993.

Concept Sharing: A Means to Improve Multi-Concept Learning , with Piew Datta, Machine Learning Conference.

``The Utility of Knowledge in Inductive Learning'', with Michael Pazzani, Machine Learning, 9, 57--94, 1992.

Utilizing Prior Concepts , with Piew Datta, Machine Learning Workshop on Bias.

``Instance-Based Learning Algorithms'', with David Aha and Marc Albert, Machine Learning, 37--66, 1991.

``SteppingStone: An Empirical and Analytic Evaluation'', with David Ruby, Proceedings of the Ninth National Conference on Artificial Intelligence, 527--531, Morgan Kaufmann, 1991.

``Machine Learning as an Experimental Science'', with Pat Langley. Readings in Machine Learning, Dietterich, T., and Shavlik, J. (eds.), 38--43, Morgan Kaufmann, 1990.

``Instance-Based Prediction of Real-Valued Attributes'', with David Aha and Marc Albert, Computational Intelligence: an International Journal, Vol 6, 3, 51--57, 1989.

``Exploring the Episodic Structure of Algebra Story Problem Solving'', with Rogers Hall, Etienne Wenger, and Chris Truxaw, Cognition and Instruction, 1989.

``Experimental Goal Regression A Method for Learning Problem Solving Heuristics'', with Bruce Porter, Machine Learning 3, 245--289, 1986.

``Differing Methodological Perspectives in Artificial Intelligence Research'', with Rogers P. Hall, Artificial Intelligence Magazine, Volume 6, Number 3, pp. 166-178, August 1985.

Professional Activities:

Reviewer for Bioinformatics, Machine Learning, KDD, IEEE

Scientific Advisor for Oncotech.

CEP and UCEP member.

Other Interests:

Reading, bridge, hiking.

Information and Computer Science
University of California, Irvine CA 92717-3425

Last modified: May 16, 2005