There's been some complaints that letter frequencies were not made available. Being CS-types, I expected you to be able to write the short C, C++, Java, perl, or whatever language programs to calculate this over your own sample text, e.g., email, web pages, etc. Included below is a perl program to compute letter frequencies; if you think you need digram frequencies, you can write it / extend it yourself, though you should be able to, by using common sense / observations of actual English text, guess what the more frequent letter pairs that you'd encounter.
perl -e '
while (<>) {
tr/A-Z/a-z/;
tr/a-z//cd;
@chars=split / */,$_;
for $letter (@chars) {
$freq{$letter}++;
}
}
for $letter (keys %freq) {
print "$letter: ",$freq{$letter},"\n";
} '
bsy+cse190@cse.ucsd.edu, last updated
