Making human place knowledge digestible by computers: Designing a place-based GIS (DP170100109)

This project aims to deliver the fundamental computational methods to capture, model, process, and interact on human place knowledge. People understand their environment in a very different way to computers: They think of places and their relations, while computers use coordinates and maps. People’s interaction with maps is cognitively costly and error-prone, which is becoming untenable in contexts requiring time-critical decision making. The new fundamental computational methods will enable smart human-machine interaction that intuitively is based on place knowledge. The developed methods will be evaluated in scenarios such as general search, crowd-sourced data capture, or interaction with autonomous vehicles.

Talking about Place – Tapping Human Knowledge to Enrich National Spatial Data Sets (LP100200199) (2010-2014)

Supported by PSMA, ESTA, CGNA

Placename databases (gazetteers or address files) form an essential backbone of any national spatial data infrastructure. This project aims to enhance the content and functionality of such placename databases by tapping human knowledge. It will collect large corpora of geotagged human place descriptions, and interpret them to reveal the use of placenames, the spatial extent of places, and the relations between places. Results will support data custodians (our partners PSMA, CGNA) to enrich their databases, and critical users (our partner ESTA, Victoria’s ‘000’ operator) to provide better services. They will form an essential key for facilitating free dialogs about locations, e.g., in emergency calls, in navigation, or in local web search.

Place descriptions are a common way for people to describe a location, but no current tools are smart enough to understand them. Emergency call centres are risking lives, address problems cost billions per year (USPS), and users of navigation or web services are frustrated. This project comes with a novel, interdisciplinary approach to automatically interpret human place descriptions. It will develop novel methods to capture placenames with their meaning for smarter databases and automatic interpretation procedures. The acquired knowledge will be an important step forward for Australia’s data custodians and for users. Australia’s location information industry will gain a significant advantage on a highly competitive global market.


Crowd-Sourcing Human Knowledge on Spatial Semantics of Place Names

Supported by Institute for a Broadband-Enabled Society (IBES), 2010

Learning from the Crowds: A mobile location-based game engaged participants to ‘Tell us where they were’. People visited the game website from their mobile, confirmed where their phone had located them, and typed in a description of where they were – just how they would describe that place to a friend. Prizes were set out wor lucky participants.

We received by this game more than 2000 place descriptions. This information will be used as part of an academic research project that aims to discover how we talk about Melbourne, or any place in Victoria. These descriptions are now investigated to develop better web searching, mapping and navigation systems, and even emergency services.