New tool ahead for language learning

Learning a new language could soon be as easy as picking up a newspaper.

Waikato University PhD student Michael Walmsley is working on a project which will help language learners build their foreign vocabulary by reading texts online where some of the words have been replaced with words in their target language.

His idea, still in its infancy, has seen him awarded a $98,000 government-funded Top Achiever's Doctoral Scholarship.

The software engineering student will spend the next three years researching ways to tap into existing online resources, such as Wikipedia and the Wiktionary, to create suitable reading texts for language learners.

Both online resources come in around 170 languages.

Mr Walmsley hopes to develop software to use them to automatically create suitable texts.

At this stage he is focusing on Japanese and Spanish with the hope to one day bring in te reo.

"The goal is to make learning a language fit into people's busy schedules," he said.

Mr Walmsley, who spent two years volunteering in Japan as a missionary, said the idea came to him through his own experiences.

He learnt to speak basic Japanese while there, but was now taking extra papers to increase his vocabulary.

"I can always find time to read the newspaper to keep up with current affairs, but I can't do that in Japanese it's just too time-consuming.

"This way, busy people can squeeze in language study by combining it with their general reading."

He also wants to develop cellphone games that people can use to reinforce the vocabulary they are learning. "Just reading the word once or twice won't be enough."

Mr Walmsley, a former Church College student, is one of five Waikato University students to receive a Top Achiever's Doctoral Scholarship this year. Another is PhD student Sam Sarjant, who has been awarded $93,000 to research ways to teach a computer to think.

The former Hauraki Plains College student will investigate ways to use reinforcement learning, which involves rewards, in a relational environment.

Computer scientists currently use reinforcement learning to train computers to play games, such as chess or backgammon.

Last year Sarjant came fifth in an international competition for creating the best computer learning strategies for the game of tetris.

Now he wants to teach computers to think outside the confines of a game. The ultimate goal will see computers operating in environments humans find difficult, such as space exploration, mining or nuclear waste clean-up operations.

Other Waikato winners are Naomi Simmonds, $97,000, Simon Ware, $92,000, and Leon Henderson, $253,000.

- © Fairfax NZ News