Text Analytics

Text Analytics: Community Question Answering


Lecture: Thursday, 13:30-15:10, S103/25 and S320/04

The first class will be held on April 20th 2017

Additional material will be distributed via the Moodle eLearning platform. The required passcode will be announced during the first lecture.

Course Content

Automatic question answering (QA) deals with automatically answering questions that are formulated by humans in natural language. This research field is closely related to natural language processing (NLP) as well as information retrieval (IR). Simple questions can usually be answered with a fact (“What is the weight of an iPhone 7?”) whereas more complex non-factoid questions often require more elaborate answer texts (“What do you think about the iPhone 7?”).

Since answers for these complex non-factoid questions cannot be found in knowledge bases, factoid QA systems and most digital assistants do not provide answers to such questions.

On the other hand, Community Question Answering (CQA) platforms such as StackExchange or GuteFrage.net explicitly allow users to ask complex questions, where answers are provided by other users of the same platform. In the past years, CQA archives have accumulated a large number of questions and answers, forming a valuable source of information.

Can we make use of this information in order to automatically answer new complex questions? This seminar will explore this idea and investigate several different related research areas. We will introduce the fundamentals of NLP and IR in the context of CQA, and review in depth the latest research in this area.


The first sessions will feature introductory lectures on the relevant NLP and IR concepts in CQA.

The timetable is as follows:

20.04.2017 Kick-off (S103/25)

27.04.2017 Information Retrieval in CQA (S103/25)

04.05.2017 Deep Learning in CQA (S103/25)

11.05.2017 Hands on session (S103/25)

18.05.2017 -

01.06.2017 Student Presentations: Question Retrieval (S320/04)

  • David Abuladze (Title: A Generalized Framework of Exploring Category Information for Question Retrieval in Community Question Answer Archives)
  • Johanna Heinz (Title: That’s Not My Question: Learning to Weight Unmatched Terms in CQA Vertical Search)
  • Jonas Kapitzke: (Title: Improving question retrieval in community question answering using world knowledge)

08.06.2017 Student Presentations: Question Retrieval (S103/25)

  • Kevin Nicolas Mayer (Title: Detecting Duplicate Posts in Programming QA Communities via Latent Semantics and Association Rules)
  • Maximillian Kircher (Title: Concept Embedded Convolutional Semantic Model for Question Retrieval)

22.06.2017 Student Presentations: Answer Selection and Re-Ranking (S320/04)

  • Martina Kettenbach (Title: Improved Representation Learning for Question Answer Matching)
  • Michael Örtl (Title: If You Can't Beat Them Join Them: Handcrafted Features Complement Neural Nets for Non-Factoid Answer Reranking)
  • Jana Vatter: (Title: What Is Not in the Bag of Words for Why-QA? )

29.06.2017 Student Presentations: Comment Classification (S320/04)

  • Claas Alexander Völker (Title: Attentive Interactive Neural Networks for Answer Selection in Community Question Answering)
  • Max Eichler (Title: Hand in Glove: Deep Feature Fusion Network for Answer Quality Prediction in Community Question Answering)
  • Susi Lee: (Title: Joint Learning with Global Inference for Comment Classification in Community Question Answering)

06.07.2017 Student Presentations: Answer Summarization (S320/04)

  • Johannes Heilmann: (Title: Metadata-aware measures for answer summarization in community question answering)

13.07.2017 Student Presentations (S103/25)

  • Vladyslav Yushchenko (Title: Novelty based Ranking of Human Answers for Community Questions)
  • Güven Gökdemir (Title: Supporting Human Answers for Advice-Seeking Questions in CQA Sites)
  • Andreas Schäffler: (Title: Multi-Column Convolutional Neural Networks with Causality-Attention for Why-Question Answering)

20.07.2017 Wrap-up (S103/25)


Papers for background reading and ideas for students will be presented during the seminar.

Teaching Staff

Andreas Rücklé (Please contact by e-mail for an appointment)

Prof. Dr. Iryna Gurevych