Tell Us Where You Are

Tell Us Where is the home of place-related research at the University of Melbourne, with partners world-wide. We study human place descriptions in order to enrich the human interaction with spatial technologies.

Place descriptions are a common way for people to describe a location, but no current software tools are smart enough to understand them. This way, emergency call centres are risking lives, postal services are wasting billions of dollars per year by addressing problems, and users of navigation or web services are frustrated about restrictive interfaces or prolonged search.

This research will develop novel, interdisciplinary approaches to automatically interpret human place descriptions. It will develop methods to capture placenames with their meaning — their true location — for smarter databases and automatic interpretation procedures. The acquired knowledge will be an important step forward for data custodians and for service users.

Congratulations Marie!

I am pleased to publish a post about Marie’s new paper. Here is the short summary of the new publication.

This paper demonstrates a framework of processes for identifying potential witnesses of events from evidence they post to social media. The research defines original evidence models for micro-blog content sources, the relative uncertainty of different evidence types, and models for testing evidence by combination. Methods to filter and extract evidence using automated and semi-automated means are demonstrated using a Twitter case study event. Further, an implementation to test extracted evidence using Dempster Shafer Theory of Evidence are presented. The results indicate that the inclusion of evidence from micro-blog text and linked image content can increase the number of micro-bloggers identified at events, in comparison to the number of microbloggers identified from geotags alone. Additionally, the number of micro-bloggers that can be tested for evidence corroboration or conflict, is increased by incorporating evidence identified in their posting history.

Testing the event witnessing status of micro-bloggers from evidence in their micro-blogs, Marie Truelove, Maria Vasardani, Stephan Winter, PLOS ONE, accepted 2 December 2017.

Our Hackfest workshop 2017 was held on 23 and 24 November at the University of Melbourne. It was an interesting and productive meeting, where RHD students from the project had the opportunity to work collaboratively on creating place graphs from maps. The idea was selected on the first day out of several promising ideas. Later in the evening, the algorithm was designed to take a selected map area as input and produce a place graph as output. Finally, on the second day, the implementation was completed, and a demo was successfully created at the end of the Hackfest.

Hackfest Demo

On Open Day at the University of Melbourne, 20 August 2017, spatial information researchers had set up a location-based game for all visitors. This game, a sort of scavenger hunt, asks people to find a marked location on campus, and provide a place description good enough to be found by their friends or family members. The collected place descriptions help the researchers create navigation systems that communicate like people do. While the participants on Open Day were awarded with attractive prizes, the game continues to be available at https://placegame.eng.unimelb.edu.au/, so feel free to participate in a volunteering spirit! It is currently set up to be played on the Parkville campus.

The Australian Research Council funds a new Discovery Project (DP170100109): Making human place knowledge digestible by computers. This project will run 2017-2019. The chief investigators are Stephan Winter, Tim Baldwin, Jochen Renz (ANU), Martin Tomko, Maria Vasardani, and Werner Kuhn (UCSB). We will report about progresses on this website.

Here is a video that summarizes some of the work we have been doing toward “smarter” spatial services.

Our last “Talking about Place” Annual Workshop will take place on Friday 21 November 2014, at the University of Melbourne. This is an opportunity for academic and industry team members alike to see the products of our research over the last year, and discuss and reflect on the project overall. Please find the event details below. We are looking forward to productive workshop!

Special Issue on Computational Models of Place 

Guest Editors: 

Simon Scheider (University of Münster, Germany), Maria Vasardani (University of Melbourne, Australia), Chris Jones (Cardiff University, UK), Ross Purves (University of Zurich, Switzerland), and Stephan Winter (University of Melbourne, Australia)

Aim and topic 

Even though places are immensely useful referents for geocoding and interlinking other information, e.g., in terms of gazetteers, place related information from the web, social media, or common language often still needs to be generated, linked and curated in a manual and time consuming fashion. This problem has become increasingly pressing in the age of Big Data, where the generation, provenance, curation and quality of place related data becomes uncontrollable and does not scale with the growth of other data in need of georeferencing.

This special issue aims to present cutting edge research on approaching computational models of place from various related disciplines, according to the spirit and scope of the journal, such as computing, cognition, language, artificial intelligence or geography.

Submitted papers will be reviewed to the usual standards of the journal. Instructions for submitting manuscripts can be found on the journal’s website (http://www.tandfonline.com/loi/hscc20). Submissions should not exceed 6000 words.

Important dates: 

Submission due: 28 February 2014

Acceptance Notification: 5 May 2014

Revisions by 1 August 2014

Our Annual “Talking about Place” Workshop will take place this Friday 29 November, at the University of Melbourne. This is an opportunity for academic and industry team members alike to interact, discuss and brainstorm on issues related to our project. Please find the event details below. We are looking forward to productive workshop!

Two members of our academic team, namely Stephan Winter and Maria Vasardani, are co-organizers of the ACM Workshop on Computational Models of Place (CoMP 2013), in conjunction with the 21^st ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems, November 5-8, 2013, Orlando, FL, USA.

The rest of the members of the organizing committee are:

Simon Scheider, University of Muenster, Germany (main organizer)

Benjamin Adams, National Center for Ecological Analysis and Synthesis, USA

Krzysztof Janowicz, University of California, Santa Barbara, USA

Detailed information about the workshop, which builds up from past, related workshops such as last year’s Place-related Knowledge Acquisition Research workshop (P-KAR 2012), can be found at: http://stko.geog.ucsb.edu/comp2013/

Our industry partners are cordially invited to participate, if they wish, to the workshop with or without paper submissions.