Created
July 15, 2012 04:33
-
-
Save hitsujiwool/3115005 to your computer and use it in GitHub Desktop.
IBM Model 1
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
/* | |
* IBM Model 1 | |
*/ | |
(function(exports) { | |
function train(sentencePairs, iteration) { | |
iteration = iteration || 100; | |
var count, | |
total, | |
t = tFunc(), | |
allForeignWords = {}, | |
allEnglishWords = {}; | |
function tokenize(sentence) { | |
return sentence.split(' '); | |
} | |
function countFunc() { | |
var count = {}; | |
return function(f, e, val) { | |
if (typeof val === 'undefined') { | |
return (count[f] = count[f] || {})[e] || 0; | |
} else { | |
(count[f] = count[f] || {})[e] = val; | |
return val; | |
} | |
}; | |
}; | |
function totalFunc() { | |
var total = {}; | |
return function(f, val) { | |
if (typeof val === 'undefined') { | |
return total[f] || 0; | |
} else { | |
total[f] = val; | |
return val; | |
} | |
}; | |
} | |
function tFunc() { | |
var table = {}; | |
function func(f, e, val) { | |
if (typeof val === 'undefined') { | |
return (table[f] = table[f] || {})[e] || 0; | |
} else { | |
(table[f] = table[f] || {})[e] = val; | |
return val; | |
} | |
}; | |
func.get = function() { | |
return table; | |
}; | |
return func; | |
} | |
// tokenize | |
sentencePairs = sentencePairs.map(function(pair) { | |
return pair.map(function(sentence) { | |
return tokenize(sentence); | |
}); | |
}); | |
sentencePairs.forEach(function(pair) { | |
var fWords = pair[0], | |
eWords = pair[1]; | |
fWords.forEach(function(f) { | |
if (!(allForeignWords[f])) allForeignWords[f] = true; | |
}); | |
eWords.forEach(function(e) { | |
if (!(allEnglishWords[e])) allEnglishWords[e] = true; | |
}); | |
}); | |
/** | |
* Learning lexical translation Models by EM Algorithum. | |
*/ | |
// Initialize t(f, e) uniformly. | |
Object.keys(allForeignWords).forEach(function(f) { | |
Object.keys(allEnglishWords).forEach(function(e) { | |
t(f, e, 1 / Object.keys(allEnglishWords).length); | |
}); | |
}); | |
// Start iteration. | |
for (var i = 0; i < iteration; i++) { | |
count = countFunc(); | |
total = totalFunc(); | |
sentencePairs.forEach(function(pair) { | |
var fWords = pair[0], | |
eWords = pair[1], | |
sTotal = {}; | |
eWords.forEach(function(e) { | |
sTotal[e] = fWords.reduce(function(res, f) { | |
return res + t(f, e); | |
}, 0); | |
}); | |
eWords.forEach(function(e) { | |
fWords.forEach(function(f) { | |
count(f, e, count(f, e) + t(f, e) / sTotal[e]); | |
total(f, total(f) + t(f, e) / sTotal[e]); | |
}); | |
}); | |
}); | |
// Update translation table. | |
Object.keys(allForeignWords).forEach(function(f) { | |
Object.keys(allEnglishWords).forEach(function(e) { | |
t(f, e, count(f, e) / total(f)); | |
}); | |
}); | |
} | |
return t; | |
} | |
exports.train = train; | |
})('object' === typeof module ? module.exports : (this.ibmModel1 = {})); |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment