International Workshop on Place-Related Knowledge Acquisition Research

31 August, Kloster Seeon, Germany
In conjuction with Spatial Cognition 2012

Photo: Rolf Gross, 1972

Aim of the workshop

Place has become a hot topic in GIScience: place is important in human cognition and communication, and hence, is a high priority for human‐computer interaction. But place is also a challenging concept to model, reason with, and analyze in information systems, because of its fluency with context shifts, and its underspecification. This workshop invites researchers from computational linguistics, data mining, artificial intelligence, geographic information science, and related disciplines to tackle this challenge. The aim is to bring together researchers with an interest in using the concept of geographic place for building smarter services and integrating heterogeneous data.



Workshop Structure

The workshop will be structured around a challenge for the participants. Participants will have the opportunity to submit position papers that can propose new approaches to that challenge, investigate the nature of the challenge, or investigate the corresponding research questions. At the workshop, participants will work together in groups on ideas emerging from these position papers.



The Challenge

Automatic estimation of the location of things or events based on verbal or graphical desctiptions, or photographs, or a combination of them.


Imagine a geo-spatial service whose interface allows for user interaction using place-based queries, references, or descriptions. Such a service may be able to deal with verbal descriptions, such as ‘the bar at the top end of Federation Square’, or with pictorial descriptions, where users sketch the location of said bar.

For true interaction the service needs to be able to ‘understand’ these kinds of descriptions in both ways, for interpreting input and for producing output. It should be able to estimate the location of a feature (in the real world) based on these descriptions, and to produce relevant descriptions of locations of features in response to user input.

This requires the modeling of ‘place’ and place-based representation mechanisms suitable for computational systems (including among others spatial relationships and varying levels of granularity). The interpretation of such descriptions likely also requires making use of various data sources and their integration, such as different independent geographical datasets (e.g., OpenStreetMap, Wikimapia, or Ordnance Survey or PSMA data), or a combination of such datasets and other media (verbal descriptions, photographs, videos). Such a service may also try to learn from its users, for example, by incorporating their knowledge about a place through methods of user-generated content or user feedback mechanisms.

The ultimate goal may be to gain systems and services that can understand and use ‘place’ in a way we humans do.



Position Papers

We invite position papers that relate to some part, or even all of this challenge. You may propose new approaches to address any (or all) of the issues involved in developing such a service. Or you may investigate the nature of the challenge and its issues; or investigate the corresponding research questions. In short, your position paper should stimulate and drive discussion at the workshop!

Submissions are now closed. We thank all the authors that submitted their work for consideration to P-KAR 2012.



Workshop Proceedings and Report

The accepted papers, individually or within the complete proceedings volume, can be accessed online at CEUR Workshop Proceedings (CEUR-WS.org) (P-KAR 2012 Proceedings). Information on how to cite the papers can be found at the end of the hosting cite's information document.



The Report from the workshop with results from the break-out sessions can be downloaded here .



Important Dates



Accepted Papers (numbering used in program below)

1. Component-wise Annotation and Analysis of Informal Place Descriptions
Igor Tytyk and Timothy Baldwin
2. Classification of Localization Utterances using a Spatial Ontology
Mohammad Fazleh Elahi, Hui Shi Hui, John A Bateman, Kathleen M. Eberhard and Matthias Scheutz
3. Representing Vague Places : Determining a Suitable Method
Mohammed Imaduddin Humayun and Angela Schwering
4. Reasoning About Large Places
Bernd Krieg-Brückner and Hui Shi
5. Intuitive and Natural Interfaces for Geospatial Data Classification
Falko Schmid, Oliver Kutz, Lutz Frommberger, Till Mossakowski, Tomi Kauppinen and Chunyuan Cai
6. Conversational Natural Language interaction for Place-related Knowledge Acquisition
Srinivasan Janarthanam, Oliver Lemon, Xingkun Liu, Phil Bartie, William Mackaness, Tiphaine Dalmas and Jana Goetze
7. From Pattern Recognition to Place Identification
Sven Eberhardt, Tobias Kluth, Christoph Zetzsche and Kerstin Schill
8. Identifying Touristic Places
Dominik Kremer and Christoph Schlieder



Workshop Program (Fri 31.08.2012)

  • 08:30 - 10:00: Setting the scene
    (08:30 - 09:15) Keynote speaker - Ross Purves
    (09:15 - 10:00) Position statements
  • 10:00 - 10:30: Coffee break
  • 10:30 - 12:00: Paper presentations (Session A)
  • (10:30 - 10:45) Paper 1
    (10:45 - 11:00) Paper 2
    (11:00 - 11:15) Paper 3
    (11:15 - 11:30) Paper 4
    (11:30 - 11:45) Paper 5
  • 12:00 - 13:30: Lunch
  • 13:30 - 14:30: Paper presentations (Session B)
    (13:30 - 13:45) Paper 6
    (13:45 - 14:00) Paper 7
    (14:00 - 14:15) Paper 8
  • 14:30 - 15:00: Coffee break
  • 15:00 - 16:30: Break outs session
  • 16:30 - 16:50: Coffee break
  • 16:50 - 18:00: Plenum
  • Reporting back - Concluding
  • 19:00 - 20:30: Dinner



Organizing Committee



Program Committee


Tim Baldwin (Melbourne, AU) Oliver Lemon (Heriot Watt University, UK)
Lawrence Cavedon (Melbourne, AU) Ross Purves (Zurich, CH)
Matt Duckham (Melbourne, AU) Simon Scheider (Münster, DE)
Chris Jones (Cardiff, UK) Lesley Stirling (Melbourne, AU)
Allison Kealy (Melbourne, AU) Stephan Winter (Melbourne, AU)
Peter Kiefer (Zurich, CH) Antonio Krueger (Saarbrücken, DE)




For any questions, please contact Maria Vasardani