|
Germanic Lexicon Project
Message Board
|
|
|
Author: Keith Briggs
Date: 2004-11-09 10:59:49
Subject: Re: Probabilistic correction
More tests on language recognition, now training on corpora of about 1.2MB of Latin, 3MB of modern English, 60KB (can't find more easily downloadable) of OE. Here a lower score means a better match.
> db="dbacl -c english -c latin -c oe -n"
> echo "gehelmian ; p. ode, ede; pp. od, ed" | ${db}
english 189.31 latin 237.54 oe 226.42
> echo "To cover with a helmet, crown;" | ${db}
english 91.99 latin 162.64 oe 166.48
> echo "galeSre, coronare" | ${db}
english 74.01 latin 74.29 oe 75.90
> echo "galeare, coronare" | ${db}
english 65.63 latin 67.22 oe 67.06
> echo "coronare" | ${db}
english 33.38 latin 31.26 oe 36.14
> echo ":-- ÐÚ gehelmodest us" | ${db}
english 121.06 latin 134.70 oe 110.85
> echo "corSnasti nos," | ${db}
english 79.27 latin 83.71 oe 98.64
> echo "coronasti nos," | ${db}
english 72.42 latin 71.94 oe 93.14
> echo "Ps. Spl. 5, 15. Of wuldre and weorþmynt ðú" | ${db}
english 205.14 latin 276.25 oe 211.67
> echo "gehelmedest hine" | ${db}
english 73.33 latin 81.37 oe 58.58
> echo "de gloria et hondre coronasti eum," | ${db}
english 169.94 latin 126.04 oe 180.23
> echo "de gloria et honore coronasti eum," | ${db}
english 168.37 latin 108.95 oe 180.77
> echo "Ps. Spl. T. 8, 6. Gehelmod " | ${db}
english 95.06 latin 114.12 oe 106.56
> echo "galeatus," | ${db}
english 44.03 latin 48.69 oe 50.74
> echo "Aelfc. Gr. 43; Som. 45, II." | ${db}
english 102.04 latin 101.88 oe 95.12
> echo "[Laym, i-helmed : O. H. Ger. gehelmot.]" | ${db}
english 173.91 latin 231.32 oe 207.33