Created
January 31, 2013 14:48
-
-
Save adrianseeley/4683335 to your computer and use it in GitHub Desktop.
JS ngram + ngram mash
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
function ngram(string) | |
{ | |
string = string.toString().toLowerCase(); | |
var valid = 'abcdefghijklmnopqrstuvwxyz '; for(var s = 0; s < string.length; s++) if(valid.indexOf(string[s]) == -1) { string = string.slice(0, s) + string.slice(s + 1); s--; } | |
string = string.split(' '); | |
var ngrams = []; | |
var to_gram = ngram_mash(string); | |
to_gram.push(string); | |
for(var g = 0; g < to_gram.length; g++) | |
{ | |
for(var l = 0; l < to_gram[g].length; l++) | |
{ | |
for(var r = l; r < to_gram[g].length; r++) | |
{ | |
var ngram = to_gram[g][l]; | |
for(var lr = l + 1; lr <= r; lr++) ngram += ' ' + to_gram[g][lr]; | |
if(ngrams.indexOf(ngram) == -1) ngrams.push(ngram); | |
} | |
} | |
} | |
console.log(ngrams); | |
} | |
function ngram_mash(split_string) | |
{ | |
var done = []; | |
var todo = []; | |
todo.push(split_string); | |
while(todo.length > 0) | |
{ | |
var current = todo[0]; todo.splice(0, 1); | |
for(var m = 0; m < current.length - 1; m++) | |
{ | |
var building = []; | |
for(var c = 0; c < current.length; c++) | |
{ | |
if(m == c) { building.push(current[c] + current[c + 1]); c++; } | |
else building.push(current[c]); | |
} | |
if(todo.indexOf(building) == -1) todo.push(building); | |
} | |
if(done.indexOf(current) == -1) done.push(current); | |
} | |
return done; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment