4 ms·

Most sources for machine translation from Arabic-English pairs are data from US Army / anti-terrorist data. So the data are learned to translate terrorist mess

Most sources for machine translation from Arabic-English pairs are data from US Army / anti-terrorist data. So the data are learned to translate terrorist messages very well, as well as army terms. However, it doesn't translate day-to-day use that well. Machine learning is only as good as the source data - and Arabic-English language pair specifically has data from US military complex, that loves translating terrorist messages. So translating Al Qaeda messaging will be good, translating regular message boards will be worse. The machine learns to translate everything as an attack, since that is what it knows. (I cannot give you exact sources, but I knew people from inside Google. Facebook uses Bing, but I think the data sources will be similar.)

kfrzcode 9y ago
> Most sources for machine translation from Arabic-English pairs are data from US Army / anti-terrorist data. Interesting! This is HN, where's the source? > I cannot give you exact sources I find your claim hard enough to believe but this seals the deal.
- Udik 9y ago
  It'll turn out in due time that Bin Laden was just sending the US his best wishes. Joking of course, but up to a point. There are cases of words from the "enemy" deliberately translated in their more literal sense when their sense is figurative (one for all, "death to America"- even the official translation is "down with America"). And statements by Bin Laden himself have been deliberately ignored in the parts that made more sense from a political point of view, to reduce them to a more comfortable "we hate you because you're free".
nametube 9y ago
You're going to have to back that up. This would be a clearly biased dataset.
megous 9y ago
Makes me less confident in US CT capabilities.
tgb 9y ago
Given that both Arabic and English are official languages of the UN (and in which all UN documents are written), wouldn't those documents form (a large part) of the data for training? Not that that would have much in the way of "Good morning" in it except perhaps speech transcriptions.

kfrzcode 9y ago

> Most sources for machine translation from Arabic-English pairs are data from US Army / anti-terrorist data. Interesting! This is HN, where's the source? > I cannot give you exact sources I find your claim hard enough to believe but this seals the deal.

Udik 9y ago
It'll turn out in due time that Bin Laden was just sending the US his best wishes. Joking of course, but up to a point. There are cases of words from the "enemy" deliberately translated in their more literal sense when their sense is figurative (one for all, "death to America"- even the official translation is "down with America"). And statements by Bin Laden himself have been deliberately ignored in the parts that made more sense from a political point of view, to reduce them to a more comfortable "we hate you because you're free".

nametube 9y ago

You're going to have to back that up. This would be a clearly biased dataset.

megous 9y ago

Makes me less confident in US CT capabilities.

tgb 9y ago

Given that both Arabic and English are official languages of the UN (and in which all UN documents are written), wouldn't those documents form (a large part) of the data for training? Not that that would have much in the way of "Good morning" in it except perhaps speech transcriptions.