Last active
April 25, 2018 03:50
-
-
Save sadikovi/1fd227d375f535858ac9716d2aa75687 to your computer and use it in GitHub Desktop.
Simple spell checker based on dynamic programming
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
abstract class Spelling | |
case class CorrectSpelling(word: String) extends Spelling | |
case class IncorrectSpelling(word: String, suggestions: List[String]) extends Spelling | |
case class Spellchecker(dictionary: String) { | |
private val numSuggestions = 10 | |
private val maxDistance = 5 | |
// set of valid words (replace with trie for space efficiency) | |
private val set = readDict(dictionary) | |
private val heap = new java.util.PriorityQueue[(Int, String)]( | |
numSuggestions, | |
new java.util.Comparator[(Int, String)] { | |
override def compare(a: (Int, String), b: (Int, String)): Int = { | |
if (a._1 > b._1) return -1 | |
if (a._1 < b._1) return 1 | |
return 0 | |
} | |
} | |
) | |
private def readDict(path: String): java.util.HashSet[String] = { | |
val iter = scala.io.Source.fromFile(path).getLines | |
val set = new java.util.HashSet[String]() | |
while (iter.hasNext) { | |
set.add(iter.next.toLowerCase) | |
} | |
return set | |
} | |
private def balance(heap: java.util.PriorityQueue[(Int, String)], size: Int): Unit = { | |
while (heap.size > size) { | |
heap.poll() | |
} | |
} | |
private def distance(s: String, t: String): Int = { | |
val n = s.length + 1 | |
val m = t.length + 1 | |
val dp = new Array[Array[Int]](n) | |
for (i <- 0 until n) dp(i) = new Array[Int](m) | |
for (i <- 0 until n) dp(i)(0) = i | |
for (j <- 0 until m) dp(0)(j) = j | |
for (i <- 1 until n) { | |
for (j <- 1 until m) { | |
val a = s.charAt(i-1) | |
val b = t.charAt(j-1) | |
dp(i)(j) = dp(i-1)(j-1) + (if (a == b) 0 else 1) | |
dp(i)(j) = Math.min(dp(i)(j), dp(i)(j-1) + 1) | |
dp(i)(j) = Math.min(dp(i)(j), dp(i-1)(j) + 1) | |
} | |
} | |
return dp(n-1)(m-1) | |
} | |
def check(word: String): Spelling = { | |
val target = word.toLowerCase | |
if (set.contains(target)) return CorrectSpelling(target) | |
heap.clear() | |
val iter = set.iterator() | |
while (iter.hasNext) { | |
val s = iter.next | |
val edits = distance(s, target) | |
if (edits <= maxDistance) { | |
heap.add((edits, s)) | |
balance(heap, numSuggestions) | |
} | |
} | |
// extract suggestions and build result | |
val suggestions = new Array[String](heap.size) | |
var i = heap.size - 1 | |
while (!heap.isEmpty) { | |
suggestions(i) = heap.poll()._2 | |
i -= 1 | |
} | |
return IncorrectSpelling(target, suggestions.toList) | |
} | |
} | |
val a = Spellchecker("words.txt") | |
a.check("deleete") | |
// res5: Spelling = IncorrectSpelling(deleete,List(delete, delate, desuete, dolente, delegate, deleted, decrete, | |
// melete, celeste, detente)) | |
a.check("deleteed") | |
// res11: Spelling = IncorrectSpelling(deleteed,List(deleted, depleted, delete, deletes, deleter, delated, deleaded, | |
// deleaved, deedeed, deletery)) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
We can optimize suggestions in several ways, e.g. short-circuiting some of the words and using global dp matrix for all solutions.