As a freelance journalist at Haymarket Media:
When he found himself staring at a screenful of irrelevant headlines in search of the few news stories of interest, Artificial Intelligence enthusiast Alex North decided that it was time to delegate the task to a machine.
Thus was born tiinker, which was launched this week by North's start-up company, Deep Grey Labs, as an intelligent news aggregator that learns the interests of individual users and selects stories tailored to each individual.
"The amount of information available on the internet is growing so quickly it is impossible to keep up without a way to weed out the chaff, so to speak," said Oleg Sushkov, co-founder of Deep Grey Labs.
"tiinker came about because we saw a need for a good news filter due to the huge number of news articles and blogs being published every day - many more than any person could reasonable keep up with."
Targeted at the 'average' Internet user who reads news online, tiinker joins an array of online news aggregation tools such as Slashdot, Digg, and Google News.
But while these existing sites either employ a group of users or editors to choose the news stories that are centrally displayed, tiinker has been designed to use machine learning and artificial intelligence techniques to tailor news feeds to individual users' preferences.
"tiinker is aimed at people who want something to just work for them automatically, not people who want absolute control," North said.
"People who might usually visit the websites of a few newspapers and blogs can come to tiinker instead and have stories picked for them, and from a far wider range of sources than they could cover otherwise."
For the past year, tiinker has been a full-time venture for Sushkov and North, who are both recent graduates of the school of Computer Science and Engineering at the University of New South Wales.
The intelligent news aggregator was one of three ideas that the pair thought up, with the aim of applying their academic knowledge in a real world product.
"There are a number of machine learning and AI [Artificial Intelligence] techniques in the backend, tuned and combined in a way we believe to be unique," North said. "Our studies in AI at uni were a good preparation for developing this - we knew what to try first and how to make it work well."
"Deep Grey Labs was started with a mission of applying research to making products and services for people," he said. "Too much research is locked away in academia, where no-one is interested in building a product out of it."
But while a bulk of Deep Grey Labs' development has been framed by Sushkov's and North's studies at the University of New South Wales, the start-up has received little financial and technical support from the university.
"Apart from casual tutoring work and the support of friends and colleagues, we've had little real support from the UNSW; IP [Intellectual property] rules make it very difficult for us to collaborate," North said.
So what's next for the boys at Deep Grey Labs?
"Hopefully making some money so we can eat," North said.
"If tiinker and Deep Grey Labs are successful we'd like to expand a little and take on another problem sometime."
"We're aiming to have a profitable business so that we can go on and apply AI and machine learning research to yet more services to help people."
