/// <summary>
/// Compares two strings for similarity.
/// </summary>
/// <param name="firstWord">The first string to compare.</param>
/// <param name="secondWord">The second string to compare.</param>
/// <param name="threshold">Any similarity score at or above this value will be considered similar.</param>
/// <param name="algorithm">Specify the algorithm to use for comparing <paramref name="firstWord"/> and <paramref name="secondWord"/>. The default
/// algorithm is <see cref="SimMetricAlgorithm.Levenstein"/>.</param>
/// <returns>`true` if the similarity score is equal or above <paramref name="threshold"/>. Otherwise, `false`.</returns>
/// <remarks><![CDATA[
/// The following code demonstrates how to filter a list of strings based on the degree of similarity:
/// ```C#
/// string word = "fooler"
/// var list = new List<string>() { "fowler", "fish", "crawler" };
/// var filtered = List<string>();
/// foreach (string item in list)
/// {
/// if (item.NearEquals(word))
/// filtered.Add(item);
/// }
/// Console.WriteLine("You typed '{0}'. Did you mean: {1}", word, filtered);
/// ```
/// ]]></remarks>
public static bool NearEquals(this string firstWord, string secondWord, double threshold, SimMetricAlgorithm algorithm)
{
IStringMetric sim = StringMetricFactory.FromAlgorithm(algorithm);
double diff = sim.GetSimilarity(firstWord, secondWord);
return((diff < threshold) ? true : false);
}
/// <summary>
/// Constructor for use by the serializer, you should be using this otherwise.
/// </summary>
/// <param name="dictionary"></param>
/// <param name="wordlist"></param>
internal SymmetricPredictor(Dictionary<string, List<int>> dictionary, List<string> wordlist, IStringMetric stringMetric)
{
_stringSimilarityProvider = stringMetric;
Dictionary = dictionary;
Wordlist = wordlist;
}
/// <summary>
/// Will process a list of words into the Dictionary and WordList
/// </summary>
/// <param name="strings">The words in the corpus.</param>
public SymmetricPredictor(IEnumerable<string> strings, IStringMetric stringMetric)
{
_stringSimilarityProvider = stringMetric;
foreach (var key in strings)
{
CreateSymmetricEntry(key);
}
Wordlist.TrimExcess();
}
/// <inheritdoc cref="IStringMatcher.IsMatch"/>
public double IsMatch(string input)
{
IStringMetric m = GetStringMetricType();
return(MatchBehaviourHelper.Convert(MatchBehaviour, MatchScores.ToScore(_patterns.Select(p => m.GetSimilarity(p, input)))));
}
internal SpellCheck(IEnumerable <string> words, IStringMetric metric) : base(words)
{
StringSimilarityProvider = metric;
}
public SpellCheck(int terminalCount, char[] characters, int rootNodeIndex, int[] firstChildForNode, int[] edges, ushort[] edgeCharacter, IStringMetric metric)
: base(terminalCount, characters, rootNodeIndex, firstChildForNode, edges, edgeCharacter)
{
StringSimilarityProvider = metric;
}
public void AddArbitrationMetric(IStringMetric arbitrationMetric) {
metricsForArbitration.Add(arbitrationMetric);
}
public static double CalcMetric(this Image image, string title, ITokenizer tokenizer, IStringMetric metric)
{
var imageTitle = string.Join(" ", tokenizer.Tokenize(image.Title));
var similarity = metric.GetSimilarity(title.ToLower(), imageTitle.ToLower());
return(similarity);
}
