The most read New Atlas stories of 2017

Here at New Atlas we cover a wide variety of topics, so it’s always interesting to arrive at the end of another year and see what most struck a chord with readers over the past 12 months. This year our top science and technology news stories ranged from ancient death traps and otherworldly weapons to anti-aging breakthroughs and mega-aircraft. Read on to see what captured the imagination of you, our readers, in 2017.

more…

Missed this one

How a computer sees history after “reading” 35 million news stories

So far, humans have relied on the written word to record what we know as history. When artificial intelligence researchers ran billions of those words from decades of news coverage through an automated analysis, however, even more patterns and insights were revealed

more….

Weren’t you fiddling with this stuff recently Mr Moll?

https://newatlas.com/britain-tallest-mountain-antarctica/52569/

“Researchers have discovered that Britain’s highest mountain isn’t in Britain. A recent cartographic survey by the British Antarctic Survey (BAS) has found that Mount Hope, at an elevation of 3,239 m (10,654 ft) above sea level, is not only the highest mountain in the British Antarctic Territory (BAT), but in any British territory. The survey, which incorporates new satellite data shows that Mount Hope is 337 m (1,106 ft) taller than previously thought, making it 55 m (180 ft) higher than Mount Jackson, the now second highest in the territory at 3,184 m (10,444 ft).”

What a find – zircons around a third as old again as the planet!

https://newatlas.com/mauritia-supercontinent-gondwana/47704/

“On site in Mauritius: Lewis Ashwal studying volcanic rocks that contain six billion-year-old zircon grains (Credit: Susan Webb/Wits University)”

Seriously though. This is very sloppy science journalism.

Tau.Neutrino said:

Missed this one

How a computer sees history after “reading” 35 million news stories

So far, humans have relied on the written word to record what we know as history. When artificial intelligence researchers ran billions of those words from decades of news coverage through an automated analysis, however, even more patterns and insights were revealed

more….

You are aware how abysmal the text recognition of scanned documents is, I take it. I bet that correct text recognition would double the accuracy of the analysis.

mollwollfumble said:

Tau.Neutrino said:

Missed this one

How a computer sees history after “reading” 35 million news stories

So far, humans have relied on the written word to record what we know as history. When artificial intelligence researchers ran billions of those words from decades of news coverage through an automated analysis, however, even more patterns and insights were revealed

more….

You are aware how abysmal the text recognition of scanned documents is, I take it. I bet that correct text recognition would double the accuracy of the analysis.

Really? I thought things had improved markedly in the last ten years or so.

Michael V said:

mollwollfumble said:

Tau.Neutrino said:

Missed this one

How a computer sees history after “reading” 35 million news stories

So far, humans have relied on the written word to record what we know as history. When artificial intelligence researchers ran billions of those words from decades of news coverage through an automated analysis, however, even more patterns and insights were revealed

more….

You are aware how abysmal the text recognition of scanned documents is, I take it. I bet that correct text recognition would double the accuracy of the analysis.

Really? I thought things had improved markedly in the last ten years or so.

Not on any scanned text website (such as Trove and scanned science papers) that I’ve seen.

mollwollfumble said:

Michael V said:

mollwollfumble said:

You are aware how abysmal the text recognition of scanned documents is, I take it. I bet that correct text recognition would double the accuracy of the analysis.

Really? I thought things had improved markedly in the last ten years or so.

Not on any scanned text website (such as Trove and scanned science papers) that I’ve seen.

Work got rid of a few accountants it is so accurate now – if you pay…