THE COLLEGE OF NEW JERSEY

Computer Science

CSC485 03: Special Topics :

Open Domain Question Answering Systems

Spring 2003


Course Goals

The goal of the information extraction (IE) is the design of systems that are capable of analyzing only the text passages which contain relevant (the system given) information. In addition, such systems do not try a comprehensive analysis of all text documents but purposely overlook the irrelevant information. Textual Question/Answering (QA) systems represent the most current trend in the information extraction from free on-line sources of text. A goal is the construction of systems, which can identify the answers to a natural-language question from a large quantity of on-line text documents. In contrast to information retrieval systems which supply a quantity of documents as a result of a simple, word-based search, QA systems are capable of  identifying the exact passages in the text of relevant documents which represent the concrete answer. In addition, there are no restrictions concerning the subject of the natural-language questions.  

Course Description

This course will discuss theory and practice of open domain question answering systems and related bibliographic information. Topics to be addressed include, but are not limited to: (i) Question answering language resources (LR) and scientific algorithm developments, (ii) Guidelines, standards, specifications, models and best practices for question answering LR, (iii) Methods, tools, and procedures for the acquisition, creation, management, access, distribution, and use of question answering LR, (iv) LR and evaluation and benchmarking of question answering systems and algorithms for tasks including (a) Advanced question analysis, (b) Answer discovery and integration, (c) Answer explanation and presentation generation, (d) Interactive question answering.

Possible joint products to be created include: (i) List of existing resources and ones under development (with planned release dates), (ii) Updates to ARDA Q&A Roadmap (www-nlpir.nist.gov/projects/duc/papers/qa.Roadmap-paper_v2.doc), (iii) List of Evaluation methods and benchmarks of question answering systems, (iv) List of unresolved research problems and/or areas in question answering, (v) Shared knowledge of research groups and efforts.

Course Prerequisites

Advanced Algorithms (CSC410) + Instructor's Permission.



About the Instructor

Dr. Miroslav Martinovic
Brief Biography

Ph.D. 1993 Belgrade University / New York University.

CS Faculty 2000-present, TCNJ

CS/Math Faculty, 1989-2000, Wagner College

CS/Math Faculty, 1983-1988, Belgrade University

Principal Scientist, 1989-present.

Research Interests
                        Question-Answering Systems
                        Natural Language Processing
                        Information Retrieval
                        Theory of Gaming

                        Computer Science Education 

                        Logic Programming
                        Expert Systems
            Sponsors
                        NSF
                        DARPA
                        NIST

                        Microsoft


E-mail Address :

          mmmartin@tcnj.edu (click to e-mail)

 




Telephone :

          (609) 771-2789.

 


Office : 

          Holman Hall 230.


 


Class Time:

Lectures 

Monday, Wednesday 9:30-10:50

At Holman Hall 128

Instructor supervised assigned work

360 minutes at students own schedule.

at Holman Hall 372


Textbooks:  

Course Main Text

 

 

Modern Information Retrieval

by : R. Baeza, B. Ribeiro, Addison Wesley,  2000.

 

ISBN 0-201-39829-X

Additional Texts

 

 

Mathematical Foundations of Information Retrieval

by: S. Dominich, Published by Kluwer Publishing, 1999.  

ISBN 0-7923-6861-4

 


Office Hours :

Monday

Tuesday

Wednesday

Thursday

Friday

 

 

 

 

 

9-9:30

 

9-9:30

9-11

 

12:30-2:30

 

12-2 (by appt.)

12-3 (by appt.)

 




Grading Policy:

Attendance, Class Participation and Effort

20%

Topic Presentation and Critique 
         Topic/Paper List (e-mail instructor for a password)
         Paper critique and presentation guidelines (e-mail instructor for a password)

35%

Final Paper/Project with Presentation and Demo 
        Project with guidelines and resources (Courtesy of C. Cardie)

(e-mail instructor for a password)

45%

 


 

 

 CSC485 03
Tentative Schedule

 

Introduction to Corpus-Based Question Answering

Week 1and 2

 

 

 

What is corpus-based Q&A ?
Evaluations of Q&A Systems : TREC
Current Approaches to Q&A
NLP & IR for Q&A Systems
Semantics in Q&A Systems

Slides (transparencies used in class)

Courtesy of : C. Monz and M. de Rijke

 


 

What's in Store for Question Answering ? Ask Jeeves

Week 2 and 3

 

 

 

"Take-home" messages when considering Q&A task
Some anectdotes and a few statistics
Prognostications

Slides (transparencies used in class)

Courtesy of : J.B. Lowe



 

Web Information Retrieval : Google's Success

Week 4

 

 

 

Paper presentation and critique.

 

Papers/Google/icde.pdf

 

Courtesy of : M. Henzinger


 

Essential Properties of Information Retrieval : NLP for IR

Week 5

 

 

 

Paper presentation and critique.

 

Papers/NLPforIR/NLP-IR.pdf

 

 

 

NLP Tools : Generic Retrieval Systems (SMART System)

Week 6

 

 

 

Paper presentation and critique with a demonstration session.

 

Papers/SMART/SmartCourse.html

 


 

NLP Tools : Part-of-Speech Taggers

Week 7

 

 

 

Paper presentation  and critique with a tagger installation and a demonstration session 

 

Paper : Papers/POSTagger/aaai94-tagger.ps
Resource directory : ~mmmartin/Information Retrieval/EricBrill'sTagger/

 


 

NLP Tools : Parsers (A Parser for English)

Week 8

 

 

 

Paper presentation  and critique with a demonstration session 

 

Papers : Papers/APParser/manual.ps, Papers/APParser/APParser.htm
Resource directory (
springfield) : /projects/mmmartin/Information Retrieval/NYU Parser/

 


 

NLP Tools : Electronic Lexicons (WordNet)

Week 9

 

 

 

Paper presentation  and critique with a demonstration session 

 

Documentation : http://www.cogsci.princeton.edu/~wn/doc.shtml 
Resource directory : ~mmmartin/www/CMSC485/Papers/WordNet/

 

 


 

Advanced Question Answering : Plenty of Challenges to Go Around

Week 10 and 11

 

 

 

AQAINT Program
Introducing ARDA
Advanced Question Ansering
     Multiple Approaches
     AQAINT Program
     Challenges from AQAINT Perspective
Some Final Thoughts

 

Slides (transparencies used in class)

 

Courtesy of : ARDA and J.D. Prange

 


 

Issues, Tasks and Program Structures to Roadmap Research in Q&A

Week 12 and 13

 

 

 

Issues in Q&A Research
    Question Classes: Need for question taxonomies
    Question Processing: Understanding, Ambiguities, Implicatures and Reformulations
    Context and Q&A
    Data Sources for Q&A
    Answer Extraction: Justification and Evaluation of Answer Correctness
    Answer Formulation
    Real Time Question Answering
    Interactive Q&A
    Advanced Reasoning for Q&A
    User Profiling for Q&A
    Collaborative Q&A
Milestones in the Program
Evaluation Framework

Slides (transparencies used in class)

Courtesy of : J.Burger, et. al.

 


 

Named Entity Recognition

Week 13

 

 

 

Paper presentation  and critique 

 

Paper resource directory : ~mmmartin/www/CMSC485/Papers/NER/

 


 

Anaphora Resolution

Week 14

 

 

 

Paper presentation  and critique 

 

Resource directory : ~mmmartin/www/CMSC485/Papers/Anaphora/

 


 

Project Presentations and Demos

Week 14

 



Attachments :


List of Papers/Topics

Topic 

Paper and Demonstration Materials

Presenter

Presentation 
Date

1.   Web Information Retrieval : Google's Success

¤    Paper : Papers/Google/icde.pdf

R. D. VonGleich

Week of 02/10

2    Essential Properties of Information Retrieval : NLP for IR

Paper : ¤ Papers/NLPforIR/NLP-IR.pdf
¤ Presentation
¤ Critique

P. Y. Ng

Week of 02/17

3.   SMART System : Paper with a demonstration session

¤    Papers/SMART/SmartCourse.html

D. S. Barber

Week of 02/24

4.   Eric Brill's Part-of-Speech Tagger : Paper with the tagger installation and demonstration 

Paper : Papers/POSTagger/aaai94-tagger.ps, Resource directory : ~mmmartin/Information Retrieval/EricBrill'sTagger/
¤ Presentation
¤ Critique

M. D. Scehovic

Week of 03/03

5.   Apple Pie Parser for English : Paper with a demonstration session

      Papers : Papers/APParser/manual.ps, Papers/APParser/APParser.htm¤ Papers/APParser/CBSSemanticParser.pdf,  ¤ Papers/APParser/SemanticParsingAIMag.pdf ,
¤ Presentation
Resource directory (on springfield) : /projects/mmmartin/Information Retrieval/NYU Parser/

K. A. Wilson, Jr.

Week of 03/17 

6.   WordNet Electronic Lexicon : Paper with a demonstration session

Documentation : http://www.cogsci.princeton.edu/~wn/doc.shtml, Resource directory : ~mmmartin/www/CMSC485/Papers/WordNet/
¤ Presentation
¤ Critique

J. N. Hankins

 Week of 03/24

7.   Named Entity Recognition

Paper Resource Directory : Papers/NER/

R. J. Wagner

Week of 04/14

8.   Anaphora Resolution

Paper Resource Directory : Papers/Anaphora/

A. Archer Waterman

Week of 04/14

8.   Anaphora Resolution

Paper Resource Directory : Papers/Anaphora/

J. M. Burger

Week of 04/14

Topic 

Paper and Demonstration Materials

Presenter

Presentation 
Date


Paper critique and presentation guidelines

 


 

Paper Critique Guidelines

 Each critique should be no more than one page long. Less than a page is OK. The purpose of a critique is not to summarize the paper; rather you should choose one or two points about the work that you found interesting.

Examples of questions that you might address are:

Your critique should be typed (single space) and should list the title of the paper and its authors at the top, along with your name.

Avoid unsupported value judgments, like ``I liked...'' or ``I disagreed with...'' If you make judgments of this sort, explain why you liked or disagreed with the point you describe.

Be sure to distinguish comments about the writing of the paper from comment about the technical content of the work.


Paper Presentation Guidelines


    Length        :    class period (60-80 minutes)

    Medium      :    PowerPoint, HTML slides, PDF slides or alike.


 

Paper Critique Presentation Guidelines


    Length        :    class time (talk of up to 40 minutes to be followed by an up to 40 minutes discussion mediated by the presenter)

    Medium      :    PowerPoint, HTML slides, PDF slides or alike.


 


Note about how the preparedness for other students presentations affects the grade

    (i)    All listed papers must be read by every student in class.

    (ii)   The discussion following the paper presentation and paper critique presentation demonstrates that the student has read the paper.

    (iii)   Student's involvement and competence in the discussion from (ii) will directly affect the "Attendance, Class Participation and Effort"'s 20% of  student's total grade for the entire course.
 


M. Martinovic

 


2003-01-16


 





CMSC485 03 Special Topics :

Open Domain Question Answering Systems

Spring 2003

Course Project

Due at the beginning of the scheduled presentation, on Monday, April 28 or Wednesday, April 30


Goal for the assignment: to gain a basic experience in the design, implementation, and evaluation of question­answering (QA) systems. The project is fairly open­ended. You are a member of a project team of two who is to implement a QA system that will operate in the standard TREC QA framework: the input to the system is a question, the output is a ranked list of five guesses for the answer. No human intervention is allowed in deriving answers.


For the assignment, we are providing a QA corpus that contains a set of questions and the expected answer(s) for each question. Since we can't make available to you the actual 9GB TREC collection used in the TREC QA studies, we will instead provide the top 20 documents retrieved by the Smart IR system (from a similarly large text collection) for each question in the corpus. Answers to each question are to be extracted from these 20 documents. Note that it is possible for some questions that none of the 20 retrieved documents contains the answer.

As noted above, the project is completely open­ended: you are free to build whatever components you'd like to include in your QA system and are free to use any publicly available software that you wish. You can even share components that you build with others in the class.

The primary caveats are that your system cannot use the answers provided and must make clear in the write­up what components you used that you did not write yourself.
Assume that your system has entered the 50­byte (short answer) QA track so all answers should be 10 or fewer words in length. In addition, the output for each question should be the following:
    question# document­id answer­text(for top­ranked guess)
    question# document­id answer­text(for second guess)
    question# document­id answer­text(for third guess)
    question# document­id answer­text(for fourth guess)
    question# document­id answer­text(for fifth guess)
The document­id refers to the document where the answer string was found. Use "nil" as the answer­text if your system finds no answer for a particular question.


What is provided :
    questions.txt: the questions (http://www.tcnj.edu/~mmmartin/CMSC485/Project/questions.txt).
            Feel free to change the format of this file if it makes automatic processing of the questions easier. Alternatively, you can use the questions as they appear in
            the "answers" file described below. In either case, you will need to keep around the question number to include as part of the answers file that your system produces.
    answers.txt: all answers found by TREC assessors for each question (http://www.tcnj.edu/~mmmartin/CMSC485/Project/answers.txt).
            The format of this file should be pretty clear. For each question, the file contains: (1) one line with the question number, (2) one line with the question, (3) list of document
            id's followed by answer strings, one per line, (4) a blank line separates the information for each question. Feel free to modify the format of this file if it's easier for your
            system to process.
    top 20 documents retrieved for each question: A gzipped­tar file with the top 20 documents retrieved for each question by Smart can be downloaded from
            http://www.tcnj.edu/~mmmartin/CMSC485/Project/top-20.tar.gz. (WinZip should open this file as well.)


Implementation hints :
    Start simple!! Select some really really dumb strategy to produce answers for each question just to make sure that you will have something to evaluate and to turn in. Only after you can do that should you proceed to something more sophisticated. It's fine to try a strategy very different from anything discussed in class. It's even fine if the system that you produce does terribly in terms of performance. You just need to be able to argue (in your write­up) why the strategy that you investigated MIGHT have worked. One possibility is to try using Lemur (http://www-2.cs.cmu.edu/~lemur/) or Smart system to implement a passage retrieval strategy for question answering. Another is to instead focus on one type of question, e.g. "who" questions, and develop a strategy specifically for that question type.


What to turn in :
    1. A description of your QA system. Enough detail should be provided so that, in theory at least, I could re­implement it. The description should explain each component in your QA system, the steps that your system takes to answer a question, any additional on­line sources of information used by the system, etc. Make clear which components of the system you built yourself vs. downloaded from elsewhere vs. got from another student in the course.
    2. The output file of answers produced by your system for the questions from the development corpus that we provided. The answers should be in the format described above.
    3. An evaluation (e.g. using the mean reciprocal rank evaluation measure) and analysis of your system's performance on the questions from the development corpus provided. How well did the system work? What worked? What didn't work? Can you say anything about which component is strongest/weakest?
    4. A detailed walk­through of what your system did to handle one question (any one) in the corpus.
    5. The output from your system for the question selected in (4) above. Enough information should be included in the output to convince me that the system is following the steps
described in (1). It is not necessary to submit your code, but I may ask to see it in cases where the system description is unclear.


Presentation guidelines :
    1. The presentation should be a 35 minutes talk.
    2. An additional 5 minutes questions session should follow the talk.
    3. The talk should include a simple demonstration which is not to exceed 15 minutes in length.