Hasso-Plattner-Institut
Prof. Dr. Felix Naumann
 

Master Seminar on Question Answering Systems

Automatic Question Answering (QA) has received researchers’ attention over the last decade. It aims to find a short answer string which exactly addresses the users' questions rather than overwhelm them with a large number of retrieved documents, which they must sort through to find the desired answer.

This goal is usually achieved using a combination of different techniques, including

  • information retrieval
  • information extraction
  • natural language processing
  • opinion mining.

In this seminar we will cover the most important components and techniques that are required for building a question answering system.

In addition, the students will build a QA system that can answer questions from "Who wants to be a millionaire?".

Organizational information

Grading Process

  • Presentation
    • preliminary presentation about one of the QA components
    • final presentation about the results of the implemented QA system
  • Implementation of the Coressponding part of a QA system for "Who wants to be a millionaire?"
  • Documentation (2 pages) 

Schedule and Materials

DateTopic(s)Presenter(s)Slides
17.10.11
  • Introduction to QA systems
  • Introduction to the WWM task
 Saeedeh Momtazi

QA-intro

WWM-intro

24.10.11
  • Assigning papers
  • Discussion on WWM
Saeedeh MomtaziQA-tasks

17.11.11

  • Answer Validation
  • Sentence Annotation: Named Entity Annotation

Matthias Kohnen

Stefan Klauck

Slides

Slides

21.11.11
  • Sentence Annotation: Dependency Parsing
  • Sentence Annotation: Semantic Role Labeling

Cindy Fähnrich

Sebastian Oergel

Slides

Slides

28.11.11
  • Discussion on WWM
05.12.11
  • Question Analysis
  • Question Classification

Oliver Richter

Sebastian Koelle

Slides

Slides

12.12.11
  • Query Construction
  • Document Retrieval
Marian Gawron
Gary Yao

Slides

Slides

02.01.12
  • Sentence Retrieval: Sentences vs. Documents
  • Sentence Retrieval: Word Relationship in SR

Patrick Schulze

Philipp Langer

Slides

Slides

09.01.12
  • Answer Extraction
  • Opinion and Polarity Classification
MinhTuan Nguyen
Nils Rethmeier

Slides

Slides

Literature

  • Answer Validation
    • Masatsugu Tonoike, Takehito Utsuro, Satoshi Sato, Answer Validation by Keyword Association, COLING Conference, 2004
    • Bernardo Magnini, Matteo Negri, Roberto Prevete, Hristo Tanev, Mining the web to validate answers to natural language questions, Data Mining Conference, 2002
    • Dan Shen, Michael Wiegand, Andreas Merkel, The Alyssa System at TREC QA 2007: Do We Need Blog06?, TREC Conference, 2007 (Section 3.6  and 3.7)
  • Sentence Annotation: Named Entity Annotation
    • David Nadeau, Satoshi Sekine, A survey of named entity recognition and classification, Linguisticae Investigaciones, 2007
    • Karën Fort, Maud Ehrmann, Adeline Nazarenko, Towards a Methodology for Named Entities Annotation, ACL workshop on Linguistic Annotation, 2009
  • Sentence Annotation: Dependency Parsing
    • Joakim Nivre, Dependency Grammar and Dependency Parsing, Journal Of The International Linguistic Association, 2005
  • Sentence Annotation: Semantic Role Labeling
    • Daniel Gildea, Danied Jurafsky, Automatic Labeling of Semantic Roles, ACL Conference, 2000
    • Daniel Gildea, Danied Jurafsky, Automatic Labeling of Semantic Roles, Journal of Computational Linguistics, 2002
  • Question Analysis
    • Deepak Ravichandran, Eduard Hovy, Learning Surface Text Patterns for a Question Answering System, ACL Conference, 2002
    • Michael Kaisser, Bonnie Webber, Question Answering based on Semantic Roles, ACL workshop on Workshop on Important Unresolved Matters, 2005
  • Question Classification
    • Xin Li, Dan Roth, Learning Question Classifiers, COLING Conference, 2002
    • Dell Zhang, Wee Sun Lee, Question Classification using Support Vector Machines, SIGIR Conference, 2003
  • Query Construction
    • Samuel Huston, W. Bruce Croft, Evaluating Verbose Query Processing Techniques, SIGIR Conference, 2010
    • Eugene Agichtein, Steve Lawrence, Luis Gravano, Learning Search Engine Specific Query Transformations for Question Answering, WWW Conference 2010
  • Document Retrieval
    • Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze, Introduction to Information Retrieval, Cambridge University Press, 2008  (Chapter 12)
  • Sentence Retrieval: Sentences vs. Documents
    • Vanessa Murdoch, Aspects of Sentence Retrieval, PhD Thesis, 2006 (Chapter 2)
    • Ronald T. Fernandez, David E. Losada, Leif A. Azzopardi, Extending the language modeling framework for sentence retrieval to include local context, Journal of Information Retrieval, 2010
  • Sentence Retrieval: Word Relationship in SR
    • Vanessa Murdock, W. Bruce Croft, A Translation Model for Sentence Retrieval, EMNLP Conference, 2005
    • Saeedeh Momtazi, Advanced Language Modeling for Sentence Retrieval and Classification in Question Answering Systems, PhD Thesis, 2010 (Chapter 5 + 8.3.1)
  • Answer Extraction
    • Dan Shen, Exploring Rich Evidence for Maximum Entropy-based Question Answering, PhD Thesis, 2008 (Chapter 6)
  • Opinion and Polarity Classification
    • Hong Yu, Vasileios Hatzivassiloglou, Towards Answering Opinion Questions: Separating Facts from Opinions and Identifying the Polarity of Opinion Sentences, EMNLP Conference, 2003

 

Further Readings:

  • Answer Validation
    • Bernardo Magnini, Matteo Negri, Roberto Prevete, Hristo Tanev, Comparing Statistical and Content-Based Techniques for Answer Validation on the Web, AI*IA, 2002
  • Question Analysis
    • Pere R. Comas, Jordi Turmo, Luis Marquez, Using Dependency Parsing and Machine Learning for Factoid Question Answering on Spoken Documents, INTERSPEECH 2010
    • Dan Shen, Mirella Lapata, Using Semantic Roles to Improve Question Answering, EMNLP Conference, 2007
  • Question Classification
    • Xin Li, Dan Roth, Learning Question Classifiers: The Role of Semantic Information, Journal of Natural Language Engineering, 2006