private static DictResult checkMultiLines(string text)
{
DictResult dictRPage = new DictResult();
dictRPage.similarity = 0;
dictRPage.text = "";
List <DictResult> dictionaryResultList = new List <DictResult>();
string[] InputFragments = text.Split('\n');
for (int k = 0; k < InputFragments.Length; k++)
{
DictResult dictRWord = checkOneLine(InputFragments[k]);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (i == dictionaryResultList.Count - 1)
{
dictRPage.text += dictionaryResultList[i].text;
}
else
{
dictRPage.text += dictionaryResultList[i].text + "\n";
}
dictRPage.similarity += dictionaryResultList[i].similarity * (double)dictionaryResultList[i].word_count;
dictRPage.word_count = +dictionaryResultList[i].word_count;
}
dictRPage.similarity /= (double)(dictRPage.word_count);
return(dictRPage);;
}
private static DictResult checkOneLine(string text)
{
DictResult dictRLine = new DictResult();
dictRLine.similarity = 0;
dictRLine.text = "";
List <DictResult> dictionaryResultList = new List <DictResult>();
string[] InputFragments = text.Split(' ');
for (int k = 0; k < InputFragments.Length; k++)
{
DictResult dictRWord = checkOneWord(InputFragments[k]);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (dictionaryResultList[i].similarity > _minWordSimilarity)
{
if (i == dictionaryResultList.Count - 1)
{
dictRLine.text += dictionaryResultList[i].text;
}
else
{
dictRLine.text += dictionaryResultList[i].text + " ";
}
dictRLine.similarity += dictionaryResultList[i].similarity;
dictRLine.word_count++;
}
}
dictRLine.similarity /= (double)(dictRLine.word_count);
return(dictRLine);
}
private static DictResult checkOneLine(string text, bool f, bool b)
{
DictResult dictRLine = new DictResult();
dictRLine.similarity = 0;
dictRLine.text = "";
List <DictResult> dictionaryResultList = new List <DictResult>();
string[] InputFragments = text.Split(' ');
for (int k = 0; k < InputFragments.Length; k++)
{
bool front = false, back = false;
if (k == 0 && f)
{
front = true;
}
else if (k == InputFragments.Length - 1 && b)
{
back = true;
}
DictResult dictRWord = checkOneWord(InputFragments[k], front, back);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (dictionaryResultList[i].similarity > _minWordSimilarity)
{
if (i == dictionaryResultList.Count - 1)
{
dictRLine.text += dictionaryResultList[i].text;
}
else
{
dictRLine.text += dictionaryResultList[i].text + " ";
}
dictRLine.similarity += dictionaryResultList[i].similarity;
dictRLine.word_count++;
}
}
dictRLine.similarity /= (double)(dictRLine.word_count);
dictRLine.text.Trim();
return(dictRLine);
}
private static DictResult checkMultiLines(string text, bool front, bool back)
{
DictResult dictRPage = new DictResult();
dictRPage.similarity = 0;
dictRPage.text = "";
List <DictResult> dictionaryResultList = new List <DictResult>();
string[] InputFragments = text.Split('\n');
for (int k = 0; k < InputFragments.Length; k++)
{
bool tfront = false, tback = false;
if (k == 0 && front)
{
tfront = true;
}
else if (k == InputFragments.Length - 1 && back)
{
tback = true;
}
DictResult dictRWord = checkOneLine(InputFragments[k], tfront, tback);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (i == dictionaryResultList.Count - 1)
{
dictRPage.text += dictionaryResultList[i].text;
}
else
{
dictRPage.text += dictionaryResultList[i].text + "\n";
}
dictRPage.similarity += dictionaryResultList[i].similarity * (double)dictionaryResultList[i].word_count;
dictRPage.word_count = +dictionaryResultList[i].word_count;
}
dictRPage.similarity /= (double)(dictRPage.word_count);
dictRPage.text.Trim();
return(dictRPage);;
}
private static DictResult checkOneWord(string text)
{
double maxSimilarity = 0;
DictResult dictR = new DictResult();
List <string> equalMinDistanceDictWordList = new List <string>();
if (text.Length == _dictionaryExactMatchStringLength)//short strings are looking for the exact match
{
maxSimilarity = findSimilarDictionaryWord(text, maxSimilarity, 0, equalMinDistanceDictWordList);
if (maxSimilarity != 1)
{
dictR.similarity = 0;
dictR.text = "";
}
else
{
dictR.similarity = maxSimilarity;
dictR.text = equalMinDistanceDictWordList[0];
}
}
else
{
for (int m = 0; m < _maxWordLength; m++)
{
maxSimilarity = findSimilarDictionaryWord(text, maxSimilarity, m, equalMinDistanceDictWordList);
}
if (maxSimilarity < _minWordSimilarity) //dictionary word not found (most similar is 1) hill vs hall = 0.333333
{
dictR.similarity = 0;
dictR.text = "";
}
else
{
dictR.similarity = maxSimilarity;
dictR.text = equalMinDistanceDictWordList[0];
}
}
return(dictR);
}
private static DictResult checkOneWord(string text, bool front, bool back)
{
double maxSimilarity = 0;
DictResult dictR = new DictResult();
List<string> equalMinDistanceDictWordList = new List<string>();
var checkExact = client.Search<Frequency>(q => q
.From(0)
.Size(100)
.Index("frequency")
.Type("frequency")
.Query(fq => fq
.Filtered(fqq => fqq
.Query(qq => qq.MatchAll())
.Filter(ff => ff
.Bool(b => b
.Must(m1 => m1.Term("word_name", text.ToLower()))
)
)
)
)
);
if (text.Length < _dictionaryExactMatchStringLength)
{
dictR.similarity = 0;
dictR.text = "";
}
else if (text.Length == _dictionaryExactMatchStringLength)//short strings are looking for the exact match
{
//maxSimilarity = findSimilarDictionaryWord(text, maxSimilarity, 0, equalMinDistanceDictWordList, true);
if (checkExact.Documents.Count()==0)
{
dictR.similarity = 0;
dictR.text = "";
}
else
{
dictR.similarity = maxSimilarity;
dictR.text = text;//NeedlemanWunschTiebreaker(equalMinDistanceDictWordList, text);
}
}
else if (checkExact.Documents.Count() > 0)
dictR.text = text;
else
{
string temp = text.ToLower();
int fuzziness = 1;
temp = temp + " ";
fuzziness += 2;
//fuzziness+=20;
if(!(char.IsLower(text[1])&&char.IsUpper(text[0])))
{
temp = " " + temp;
fuzziness += 2;
}
if (text.Length > 3)
fuzziness++;
if (text.Length > 4)
fuzziness++;
//if (text.Length > 5)
// fuzziness++;
double minsim = 1- (double) fuzziness/(double) temp.Length-.001;
var searchResults = client.Search<Frequency>(s => s
.From(0)
.Size(10000)
.Index("frequency")
.Type("frequency")
.Query(q => q
.FuzzyLikeThis(fz => fz.OnFields(w=>w.word_name).LikeText(temp).MaxQueryTerms(20000).MinimumSimilarity(minsim))
)
);
Log.WriteLine("EDIT DISTANCE: " + fuzziness+". MINSIM:" + minsim);
//dictR.similarity = maxSimilarity;
//Log.WriteLine("--------------------------------------------------------------------------------------"+text);
//equalMinDistanceDictWordList = searchResults.Documents.ToList();
var wordList = searchResults.Documents;
if (wordList.Count() == 1)
{
dictR.text = wordList.ToList()[0].word_name;// NeedlemanWunschTiebreaker(equalMinDistanceDictWordList, text);
//Log.WriteLine("%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%" + dictR.text);
}
else
dictR.text = NeedlemanWunschTiebreaker(wordList, text, front, back);
//Log.WriteLine("--------------------------------------------------------------------------------------");
}
dictR.text=dictR.text.Trim('\n','\r',' ');
return dictR;
}
private static DictResult checkOneLine(string text, bool f, bool b)
{
DictResult dictRLine = new DictResult();
dictRLine.similarity = 0;
dictRLine.text = "";
List<DictResult> dictionaryResultList = new List<DictResult>();
string[] InputFragments = text.Split(' ');
for (int k = 0; k < InputFragments.Length; k++)
{
bool front = false, back = false;
if (k == 0 && f)
front = true;
else if (k == InputFragments.Length - 1 && b)
back = true;
DictResult dictRWord = checkOneWord(InputFragments[k], front, back);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (dictionaryResultList[i].similarity > _minWordSimilarity)
{
if (i == dictionaryResultList.Count - 1)
dictRLine.text += dictionaryResultList[i].text;
else
dictRLine.text += dictionaryResultList[i].text + " ";
dictRLine.similarity += dictionaryResultList[i].similarity;
dictRLine.word_count++;
}
}
dictRLine.similarity /= (double)(dictRLine.word_count);
dictRLine.text.Trim();
return dictRLine;
}
private static DictResult checkMultiLines(string text, bool front, bool back)
{
DictResult dictRPage = new DictResult();
dictRPage.similarity = 0;
dictRPage.text = "";
List<DictResult> dictionaryResultList = new List<DictResult>();
string[] InputFragments = text.Split('\n');
for (int k = 0; k < InputFragments.Length; k++)
{
bool tfront = false, tback = false;
if(k==0&&front)
tfront = true;
else if (k == InputFragments.Length-1 && back)
tback= true;
DictResult dictRWord = checkOneLine(InputFragments[k], tfront, tback);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (i == dictionaryResultList.Count - 1)
dictRPage.text += dictionaryResultList[i].text;
else
dictRPage.text += dictionaryResultList[i].text + "\n";
dictRPage.similarity += dictionaryResultList[i].similarity * (double)dictionaryResultList[i].word_count;
dictRPage.word_count = +dictionaryResultList[i].word_count;
}
dictRPage.similarity /= (double)(dictRPage.word_count);
dictRPage.text.Trim();
return dictRPage; ;
}
private static DictResult checkOneWord(string text)
{
double maxSimilarity = 0;
DictResult dictR = new DictResult();
List<string> equalMinDistanceDictWordList = new List<string>();
if (text.Length == _dictionaryExactMatchStringLength)//short strings are looking for the exact match
{
maxSimilarity = findSimilarDictionaryWord(text, maxSimilarity, 0, equalMinDistanceDictWordList);
if (maxSimilarity != 1)
{
dictR.similarity = 0;
dictR.text = "";
}
else
{
dictR.similarity = maxSimilarity;
dictR.text = equalMinDistanceDictWordList[0];
}
}
else
{
for (int m = 0; m < _maxWordLength; m++)
maxSimilarity = findSimilarDictionaryWord(text, maxSimilarity, m, equalMinDistanceDictWordList);
if (maxSimilarity < _minWordSimilarity) //dictionary word not found (most similar is 1) hill vs hall = 0.333333
{
dictR.similarity = 0;
dictR.text = "";
}
else
{
dictR.similarity = maxSimilarity;
dictR.text = equalMinDistanceDictWordList[0];
}
}
return dictR;
}
private static DictResult checkOneLine(string text)
{
DictResult dictRLine = new DictResult();
dictRLine.similarity = 0;
dictRLine.text = "";
List<DictResult> dictionaryResultList = new List<DictResult>();
string[] InputFragments = text.Split(' ');
for (int k = 0; k < InputFragments.Length; k++)
{
DictResult dictRWord = checkOneWord(InputFragments[k]);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (dictionaryResultList[i].similarity > _minWordSimilarity)
{
if (i == dictionaryResultList.Count - 1)
dictRLine.text += dictionaryResultList[i].text;
else
dictRLine.text += dictionaryResultList[i].text + " ";
dictRLine.similarity += dictionaryResultList[i].similarity;
dictRLine.word_count++;
}
}
dictRLine.similarity /= (double)(dictRLine.word_count);
return dictRLine;
}
private static DictResult checkMultiLines(string text)
{
DictResult dictRPage = new DictResult();
dictRPage.similarity = 0;
dictRPage.text = "";
List<DictResult> dictionaryResultList = new List<DictResult>();
string[] InputFragments = text.Split('\n');
for (int k = 0; k < InputFragments.Length; k++)
{
DictResult dictRWord = checkOneLine(InputFragments[k]);
dictionaryResultList.Add(dictRWord);
}
for (int i = 0; i < dictionaryResultList.Count; i++)
{
if (i == dictionaryResultList.Count - 1)
dictRPage.text += dictionaryResultList[i].text;
else
dictRPage.text += dictionaryResultList[i].text + "\n";
dictRPage.similarity += dictionaryResultList[i].similarity * (double)dictionaryResultList[i].word_count;
dictRPage.word_count = +dictionaryResultList[i].word_count;
}
dictRPage.similarity /= (double)(dictRPage.word_count);
return dictRPage; ;
}
private static DictResult checkOneWord(string text, bool front, bool back)
{
double maxSimilarity = 0;
DictResult dictR = new DictResult();
List <string> equalMinDistanceDictWordList = new List <string>();
var checkExact = client.Search <Frequency>(q => q
.From(0)
.Size(100)
.Index("frequency")
.Type("frequency")
.Query(fq => fq
.Filtered(fqq => fqq
.Query(qq => qq.MatchAll())
.Filter(ff => ff
.Bool(b => b
.Must(m1 => m1.Term("word_name", text.ToLower()))
)
)
)
)
);
if (text.Length < _dictionaryExactMatchStringLength)
{
dictR.similarity = 0;
dictR.text = "";
}
else if (text.Length == _dictionaryExactMatchStringLength)//short strings are looking for the exact match
{
//maxSimilarity = findSimilarDictionaryWord(text, maxSimilarity, 0, equalMinDistanceDictWordList, true);
if (checkExact.Documents.Count() == 0)
{
dictR.similarity = 0;
dictR.text = "";
}
else
{
dictR.similarity = maxSimilarity;
dictR.text = text;//NeedlemanWunschTiebreaker(equalMinDistanceDictWordList, text);
}
}
else if (checkExact.Documents.Count() > 0)
{
dictR.text = text;
}
else
{
string temp = text.ToLower();
int fuzziness = 1;
temp = temp + " ";
fuzziness += 2;
//fuzziness+=20;
if (!(char.IsLower(text[1]) && char.IsUpper(text[0])))
{
temp = " " + temp;
fuzziness += 2;
}
if (text.Length > 3)
{
fuzziness++;
}
if (text.Length > 4)
{
fuzziness++;
}
//if (text.Length > 5)
// fuzziness++;
double minsim = 1 - (double)fuzziness / (double)temp.Length - .001;
var searchResults = client.Search <Frequency>(s => s
.From(0)
.Size(10000)
.Index("frequency")
.Type("frequency")
.Query(q => q
.FuzzyLikeThis(fz => fz.OnFields(w => w.word_name).LikeText(temp).MaxQueryTerms(20000).MinimumSimilarity(minsim))
)
);
Log.WriteLine("EDIT DISTANCE: " + fuzziness + ". MINSIM:" + minsim);
//dictR.similarity = maxSimilarity;
//Log.WriteLine("--------------------------------------------------------------------------------------"+text);
//equalMinDistanceDictWordList = searchResults.Documents.ToList();
var wordList = searchResults.Documents;
if (wordList.Count() == 1)
{
dictR.text = wordList.ToList()[0].word_name; // NeedlemanWunschTiebreaker(equalMinDistanceDictWordList, text);
//Log.WriteLine("%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%" + dictR.text);
}
else
{
dictR.text = NeedlemanWunschTiebreaker(wordList, text, front, back);
}
//Log.WriteLine("--------------------------------------------------------------------------------------");
}
dictR.text = dictR.text.Trim('\n', '\r', ' ');
return(dictR);
}