Bridging the Communication Gap

University of Melbourne researchers have developed a mobile phone game that will help ensure computers better understand the way we talk about our towns.

University of Melbourne researchers have developed a mobile phone game that will help ensure computers better understand the way we talk about our towns.

Associate Professor Stephan Winter from the Department of Geomatics at the University of Melbourne is leading a research team of linguists, computer scientists, and spatial information scientists to try and bridge the gap in the way we explain locations to our friends and how we type descriptions into mapping services such as google maps or in-car navigation services.

The first part of their project, is a mobile phone game that encourages users to log on to www.telluswhere.net and describe where they are.

“When asked to describe a location to a friend; most of us will provide vague descriptions involving landmarks and our proximity to things. Yet when asked to type a location into Google Maps or a car navigation system; users are required to type in exact addresses and specific information,” Associate Professor Winter says.

He says computers struggle to translate the simple and natural language we use to describe locations, and we need to overcome this obstacle in order to enable smarter navigation and mapping services. The game – which involves the user logging in, confirming on a map where their phone has located them, and typing in a description of where they are – will be used by researchers to gather the words we use to describe a place.

“We’re not interested really in where they are, but how they describe where they are. So we need a large number of such place descriptions to learn from,” he says.

“Improving the way computers recognise natural language is not just for greater convenience and more effective communication; but also for more serious situations such as taking calls in emergency call centres where operators are often faced with the dilemma of inputting vague data about the location of a distressed caller into an information system which only understands specific and accurate details.”