How are Fluency Fast Tracks and Most Common Word Groupings made?


#1

Traditionally, I had always thought that Fluency Fast Tracks and Most Common Word Groupings were created (automatically or not) after a language hit a certain number of sentences. However, I recently noticed that that is not always the case. For example:

German from Japanese: 29,972 sentences – no groupings/fast track
German from Arabic: 11,814 sentences – groupings/fast track available
Tagalog from English: 8,382 sentences – no groupings/fast track
Swedish from Portuguese: 6,896 sentences – groupings/fast track available

So I’m not quite sure what the logic is behind who gets the groupings/fast track and who doesn’t. Consequently, I’m curious - how does a Fluency Fast Track and Most Common Word Groupings get made? What’s the process behind it all?

(Note: This isn’t intended as an angry “where’s my feature” post, this is simply a polite inquiry from a curious user. The examples were chosen completely at random, and I have no stake in any of them.)


#2

Hi! Thanks for posting! Good question - we try to create the Fast Tracks and Most Common Words groupings once a language pairing gets to ~10,000 sentences. The process is automated, but choosing which language pairing to run the process on is not (though it probably should be at this point).

We rolled these features out slowly initially, so some language pairings still don’t have them like you listed. It seems like both the Fast Track and Most Common Words groupings are here to stay and we should probably do an audit to add them for any remaining language pairings (thanks for the reminder!). The logic up to now has been simply that we try to add the Fast Track and Most Common Words groupings when a new pairing is added, or when someone requests it for older language pairings.

They’re made by looking at the difficulty of each sentence, where the difficulty is the missing word ranking on a frequency list for the language. For the Fast Track one sentence is randomly selected per difficulty, and for the Most Common Words groupings the sentences are grouped by the given difficulty ranges.

Interested to hear what you think, if there’s anything we might add or improve, as well as if you have any further questions. In the meantime we’ll work on getting the Fast Track and Most Common Words groupings added for the language pairings you listed. Thanks again!


#3

Wow, thanks for the detailed response! It’s awesome getting a peek into the process behind it all.

The pairings I listed are actually pretty random (I’m only really interested in the Tagalog-English) but it’d be great to see more! In particular I’m interested in seeing more Most Common Word groupings, since I’ve found that’s a great way to get off the ground when first learning a language (and therefore could more easily start learning Tagalog, haha).

Thanks for all the work you’re doing, and for being so involved here on the forums!