/// <summary>
/// Perform a Fuzzy Search of the submitted team against a list of team names to find the best match.
/// </summary>
/// <param name="submittedTeam"></param>
/// <returns></returns>
public static string DoSearch(string submittedTeam)
{
double ratOberDist = 999 /*, jaccardDist = 999*/;
//string overlapCoeff = "", lcSubseqDist = "", lcSubstrDist = "", levDist = "", normLevDist = "";
//string hammDist = "", jaroWinkDist = "", sorensenDist = "", tanimotoDist = "";
// Read list of team names from CSV generated from the BigQuery Betfair dataset
List <string> apiTeams = File.ReadLines(HttpContext.Current.Server.MapPath("soccer-teams-top500.txt")).ToList();
// Check if the submitted team exactly matches any of the team names in the list and return if match found
if (apiTeams.Contains(submittedTeam))
{
return(submittedTeam);
}
// If exact match not found, then create a KeyValuePair List
var approxMatches = new List <KeyValuePair <string, double> >();
int approxMatchesFound = 0; // Set counter to 0
// Loop through each team in the apiTeams List
foreach (var apiTeam in apiTeams)
{
// Check if the IsApproxEqual method returns true
if (FuzzyMatching.IsApproxEqual(submittedTeam, apiTeam))
{
// Get the Ratcliff Obershelp Similarity
ratOberDist = FuzzyMatching.GetRatOberDistance(submittedTeam, apiTeam);
// After trial and error the above was found to be the most suitable distance measure to use
//
//overlapCoeff = FuzzyMatching.GetOverlapCoefficient(submittedTeam, apiTeam).ToString("0.00");
//lcSubseqDist = FuzzyMatching.GetLCSubsequence(submittedTeam, apiTeam).ToString();
//lcSubstrDist = FuzzyMatching.GetLCSubsring(submittedTeam, apiTeam).ToString();
//jaccardDist = FuzzyMatching.GetJaccardDistance(submittedTeam, apiTeam);
//levDist = FuzzyMatching.GetLevDistance(submittedTeam, apiTeam).ToString();
//normLevDist = FuzzyMatching.GetNormLevDistance(submittedTeam, apiTeam).ToString("0.00");
//hammDist = FuzzyMatching.GetHammingDistance(submittedTeam, apiTeam).ToString();
//jaroWinkDist = FuzzyMatching.GetJaroWinklerDistance(submittedTeam, apiTeam).ToString("0.00");
//sorensenDist = FuzzyMatching.GetSorensonDistance(submittedTeam, apiTeam).ToString("0.00");
//tanimotoDist = FuzzyMatching.GetTanimotoDistance(submittedTeam, apiTeam).ToString("0.00");
approxMatchesFound++; // Increment counter
// Store the matched team and it's Ratcliff Obershelp Similarity measure to the submitted team
approxMatches.Add(new KeyValuePair <string, double>(apiTeam, ratOberDist));
}
}
// If only one approximate match is found then return it.
if (approxMatchesFound == 1)
{
return(approxMatches[0].Key);
}
// If more than one approximate match is found then return the one with the highest Ratcliff Obershelp Similarity
else if (approxMatchesFound > 1)
{
var max = default(KeyValuePair <string, double>);
foreach (var match in approxMatches)
{
if (match.Value > max.Value)
{
max = match;
}
}
return(max.Key);
}
else
{
return("NO MATCH");
}
}
protected void Button1_Click(object sender, EventArgs e)
{
string submittedTeam = SourceTextBox.Text;
MatchResult.Text = FuzzyMatching.DoSearch(submittedTeam);
/*
* bool MatchFound = false;
*
* string approxMatch = "";
* string overlapCoeff = "";
* string lcSubseqDist = "";
* string lcSubstrDist = "";
* string levDist = "";
* string normLevDist = "";
* string hammDist = "";
* string jaroWinkDist = "";
* double jaccardDist = 999;
* double ratOberDist = 999;
* string sorensenDist = "";
* string tanimotoDist = "";
*
* List<string> apiTeams = System.IO.File.ReadLines(Server.MapPath("soccer-teams-top500.txt")).ToList();
* foreach (var apiTeam in apiTeams)
* {
* MatchResult.Text = "";
* ResultLabel.Text = "";
*
* if (apiTeam == submittedTeam)
* {
* MatchResult.Text = MatchResult.Text + submittedTeam + " : " + apiTeam + " = Match";
* MatchFound = true;
* break;
* }
* }
* if (!MatchFound) {
* var approxMatches = new List<KeyValuePair<string, double>>();
* int approxMatchesFound = 0;
*
* foreach (var apiTeam in apiTeams)
* {
* MatchResult.Text = "";
* ResultLabel.Text = "";
*
* if (FuzzyMatching.IsApproxEqual(submittedTeam, apiTeam))
* {
* // TODO: find best match - store all matches in list
* overlapCoeff = FuzzyMatching.GetOverlapCoefficient(submittedTeam, apiTeam).ToString("0.00");
* lcSubseqDist = FuzzyMatching.GetLCSubsequence(submittedTeam, apiTeam).ToString();
* lcSubstrDist = FuzzyMatching.GetLCSubsring(submittedTeam, apiTeam).ToString();
*
* levDist = FuzzyMatching.GetLevDistance(submittedTeam, apiTeam).ToString();
* normLevDist = FuzzyMatching.GetNormLevDistance(submittedTeam, apiTeam).ToString("0.00");
* hammDist = FuzzyMatching.GetHammingDistance(submittedTeam, apiTeam).ToString();
* jaroWinkDist = FuzzyMatching.GetJaroWinklerDistance(submittedTeam, apiTeam).ToString("0.00");
* jaccardDist = FuzzyMatching.GetJaccardDistance(submittedTeam, apiTeam);
* ratOberDist = FuzzyMatching.GetRatOberDistance(submittedTeam, apiTeam);
* sorensenDist = FuzzyMatching.GetSorensonDistance(submittedTeam, apiTeam).ToString("0.00");
* tanimotoDist = FuzzyMatching.GetTanimotoDistance(submittedTeam, apiTeam).ToString("0.00");
*
* approxMatchesFound++;
* approxMatches.Add(new KeyValuePair<string, double>(apiTeam, ratOberDist));
* //MatchResult.Text = approxMatches.ElementAt(approxMatchesFound-1).ToString();
*
* //approxMatch = "Is approximate match.";
* //MatchResult.Text = MatchResult.Text + submittedTeam + " : " + apiTeam + " = " + approxMatch + "<br />";
*
* //break;
* }
* else if (approxMatchesFound == 0)
* {
* approxMatch = "Not approximate match.";
* MatchResult.Text = submittedTeam + " : " + apiTeam + " = " + approxMatch + "<br />";
* }
* }
*
* var max = default(KeyValuePair<string, double>);
* foreach (var match in approxMatches)
* {
* if (match.Value > max.Value)
* max = match;
* }
* MatchResult.Text = MatchResult.Text + "<br />" + max + " : " + approxMatchesFound;
*
* ResultLabel.Text =
* "<br />Overlap coeffecient [1]: " + overlapCoeff + "<br />" +
* "<br />Longest common subsequence: " + lcSubseqDist + "<br />" +
* "<br />Longest common substring: " + lcSubstrDist + "<br />" +
* "<br /><br />Levenshtein distance [0]: " + levDist + "<br />" +
* "<br />Normalised Levenshtein distance [0]: " + normLevDist + "<br />" +
* "<br />Hamming distance [0]: " + hammDist + "<br />" +
* "<br />Jaro-Winkler distance [1]: " + jaroWinkDist + "<br />" +
* "<br />Jaccard distance [0]: " + jaccardDist + "<br />" +
* "<br />Ratcliff-Obershelp Similarity [1]: " + ratOberDist + "<br />" +
* "<br />Sorenson-Dice distance: [0]" + sorensenDist + "<br />" +
* "<br />Tanimoto coeffecient: [1]" + tanimotoDist + "<br />";
*
* }*/
}
