As I understand it, they just use an algorithm to build the probability engine, and they've just fed it many books so it could learn the natural probabilities in the English language. So while they may not get you their exact probability statistics, the idea of the program is that it can learn your own probability of words.
<br><br>I'm sure you could feed it old emails or something to give it a feel for how you write.<br><br><br><div><span class="gmail_quote">On 10/29/07, <b class="gmail_sendername">Oliver</b> <<a href="mailto:embry00s@gmail.com">
embry00s@gmail.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">Tjenare Lars!<br><br>I think a good place to find such information is the Dasher project. (
<a href="http://www.inference.phy.cam.ac.uk/dasher/" target="_blank" onclick="return top.js.OpenExtLink(window,event,this)">http://www.inference.phy.cam.ac.uk/dasher/</a> ) They use a probabalistic model to make another alternative input device, using the same sort of statistics that you need. The project is open source, so even though the stats may not be available on the site (even though they probably are) you should just need to ask for them.
<br><br>hälsningar,<br><span class="sg">Oliver<br>
</span><br>_______________________________________________<br>OpenMoko community mailing list<br><a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:community@lists.openmoko.org">community@lists.openmoko.org
</a><br><a onclick="return top.js.OpenExtLink(window,event,this)" href="http://lists.openmoko.org/mailman/listinfo/community" target="_blank">http://lists.openmoko.org/mailman/listinfo/community</a><br><br></blockquote></div>
<br>