Wondering where fossil is? This software knows!

WASHINGTON, USA: A software based on artificial intelligence can pinpoint productive fossil sites, eliminating guesswork from the process.

Traditionally, fossil hunters often could only make educated guesses as to where fossils lie. The rest lay with chance-finding the proverbial needle in a haystack.

"I don't want to say it's total luck," says Glenn Conroy, professor of physical anthropology in Arts & Sciences at the Washington University, St Louis, "but it's a combination of hard work, meticulous planning and, well, a good dose of luck".

The Great Divide Basin, a 4,000-square-mile stretch of rocky desert in Wyoming, US, is a rich site for paleontologists, yielding 50 million to 70 million-year-old early mammal fossils, according to a university statement.

Conroy and colleagues Robert Anemone and Charles Emerson, from the Western Michigan University, developed the tool based on artificial intelligence.

With information gathered from maps and satellite imagery - such as elevation, slope, terrain and other features - the artificial neural network (ANN) was 'trained' to use details of existing fossil areas to predict locations of other fossil sites in the Great Divide Basin.

"The beauty and power of neural networks lie in the fact that they are capable of learning," says Conroy, also a professor of anatomy and neurobiology at the Washington School of Medicine.

"You just need to give them a rule to deal with things they don't know."

Conroy and colleagues tested the software at the Great Divide Basin last summer.

The ANNs correctly identified 79 per cent of the area's known fossil sites and 99 per cent of the sites it tagged contained fossils.

Conroy is next planning to continue to use the software to search for early hominid fossil sites in South Africa.