public List <string> Search(string word, int tolerance = 2)
{
List <string> results = new List <string>();
if (word == null || word.Length == 0 || _root == null)
{
return(results);
}
Queue <Node> nodesToSearch = new Queue <Node>();
nodesToSearch.Enqueue(_root);
while (nodesToSearch.Count > 0)
{
Node curr = nodesToSearch.Dequeue();
int dist = FuzzyStringMatching.LevenshteinDistance(curr.Word, word);
int minDist = dist - tolerance;
int maxDist = dist + tolerance;
if (dist <= tolerance)
{
results.Add(curr.Word);
}
foreach (int key in curr.Keys.Where(key => key >= minDist && key <= maxDist))
{
nodesToSearch.Enqueue(curr[key]);
}
}
return(results);
}
/// <summary>
/// Gets the words that partially match the provided word according to the Jaro-Winkler similarity.
/// The provided word must be 3 characters or longer.
/// </summary>
/// <param name="word">Find words that are close matches to this word. Exact matches are omitted.</param>
/// <param name="minSimilarity">The minimum similarity value for a word match to be accepted, where 1 represents matching same word.</param>
/// <returns>A list of similar words.</returns>
public List <string> GetFuzzyMatchedWordsAlternate(string word, float minSimilarity = 0.7f)
{
List <string> words = new List <string>();
if (word.Length < 3 || minSimilarity <= 0f || minSimilarity >= 1f)
{
return(words);
}
foreach (var kv in _words)
{
string targetWord = kv.Key;
if (FuzzyStringMatching.JaroWinklerSimilarity(word, targetWord) >= minSimilarity)
{
words.Add(targetWord);
}
}
if (words.Contains(word))
{
words.Remove(word);
}
return(words);
}
public void AddWord(string word)
{
if (_root == null)
{
_root = new Node(word);
return;
}
Node curr = _root;
int dist = FuzzyStringMatching.LevenshteinDistance(curr.Word, word);
while (curr.ContainsKey(dist))
{
if (dist == 0)
{
return;
}
curr = curr[dist];
dist = FuzzyStringMatching.LevenshteinDistance(curr.Word, word);
}
curr.AddChild(dist, word);
}