private void button2_Click(object sender, EventArgs e)
{
GRAMCLASS gramclass = new GRAMCLASS();
if (MemoryClass.Instance.IsEmptyBigramList())
{
gramclass.BiGram2(MemoryClass.Instance.Get_Sentences_From_OldCorpus() == ""?Regex_Patterns.Instance.Get_Corpus():MemoryClass.Instance.Get_Sentences_From_OldCorpus(), true);
}
GRAMCLASS gc = new GRAMCLASS();
string currentSentence = string.Empty;
//dataGridView1.Columns.Clear();
DataTable mytable = new DataTable();
mytable.Columns.Add("Sentence", typeof(string));
mytable.Columns.Add("Probability", typeof(double));
DataRow row;
double SumProbability = 0;
for (int index_gridview_row = 0; index_gridview_row < dataGridView1.Rows.Count - 1; index_gridview_row++)
{
//foreach (string sentence in )
currentSentence = dataGridView1[0, index_gridview_row].Value.ToString();
ArrayList SS = gc.Seperator_Phase4(currentSentence, false);
double PDF = 1;
StructData_BiGram dataBigram = new StructData_BiGram();
for (int i = 1; i < SS.Count; i++)
{
dataBigram.CurrentToken = (string)SS[i].ToString().ToLower();
dataBigram.PrevToken = (string)SS[i - 1].ToString().ToLower();
if (MemoryClass.Instance.GetListBigram.Exists(s => s.CurrentToken == dataBigram.CurrentToken && s.PrevToken == dataBigram.PrevToken))
{
StructData_BiGram bigrm = MemoryClass.Instance.GetListBigram[MemoryClass.Instance.GetListBigram.FindIndex(s => s.CurrentToken == dataBigram.CurrentToken && s.PrevToken == dataBigram.PrevToken)];
PDF *= Math.Pow(bigrm.Probability, (double)1 / SS.Count);
}
else
{
PDF = 0;
}
}
row = mytable.NewRow();
row[0] = currentSentence;
row[1] = PDF; //*((double)currentSentence.Length/Max_Index);
SumProbability += PDF; // *((double)currentSentence.Length / Max_Index);
mytable.Rows.Add(row);
}
DataSet ds = new DataSet();
ds.Tables.Add(mytable);
dataGridView1.DataSource = ds.Tables[0].AsDataView();
lblAverageResult.Text = ((double)SumProbability / Regex_Patterns.Instance.GetSentenceList.Count).ToString();
}
private void button6_Click(object sender, EventArgs e)
{
string sentences = string.Empty;
foreach (string stnc in Regex_Patterns.Instance.GetSentenceList)
{
sentences += stnc;
}
DataSet ds = new DataSet();
DataTable mytable = new DataTable();
mytable.Columns.Add("Token", typeof(string));
mytable.Columns.Add("PrevToken", typeof(string));
mytable.Columns.Add("Probility", typeof(double));
mytable.Columns.Add("Count", typeof(int));
DataRow r;
double SumProbability = 0;
List <StructData_BiGram> l = new GRAMCLASS().BiGram2(sentences, false);
foreach (StructData_BiGram SD in l)
{
r = mytable.NewRow();
r[0] = SD.CurrentToken;
r[1] = SD.PrevToken;
r[2] = SD.Probability;
SumProbability += SD.Probability;
r[3] = SD.Count;
mytable.Rows.Add(r);
}
ds.Tables.Add(mytable);
dataGridView1.DataSource = ds.Tables[0].DefaultView;
lblAverageResult.Text = ((double)SumProbability / l.Count).ToString();
}
private void AutoChange_Load(object sender, EventArgs e)
{
Fill_MemoryClass();
GRAMCLASS gc = new GRAMCLASS();
DataTable table1 = new DataTable();
table1.Columns.Add("جمله", typeof(string));
DataRow row;
foreach (string sentence in Regex_Patterns.Instance.GetSentenceList)
{
//------ add to table ------
row = table1.NewRow();
row[0] = sentence;
table1.Rows.Add(row);
}
DataSet ds = new DataSet();
ds.Tables.Add(table1);
dataGridView1.DataSource = ds.Tables[0].AsDataView();
dataGridView1.AutoSizeColumnsMode = DataGridViewAutoSizeColumnsMode.Fill;
//------ show colors for change events in datagridview
lblDelete.BackColor = Color.SeaGreen;
lblInsert.BackColor = Color.DarkKhaki;
lblSwitch.BackColor = Color.Red;
lblIncorrectSpell.BackColor = Color.Silver;
//----------------Disable buttons -----------------
button2.Enabled = false;
button3.Enabled = false;
button4.Enabled = false;
}
private void button1_Click_1(object sender, EventArgs e)
{
GRAMCLASS gc = new GRAMCLASS();
string corpus = string.Empty;
Fill_MemoryClass();
openFileDialog1.Filter = "HtmFile (*.htm)|*.htm|Htmlfile (*.html)|*.html|MhtFile(*.mht)|*.mht";
DialogResult dr = openFileDialog1.ShowDialog();
if (dr == DialogResult.OK)
{
StreamReader file = new StreamReader(openFileDialog1.FileName);
corpus = file.ReadToEnd();
Form1 f1 = new Form1();
HtmlAgilityPack.HtmlDocument document = new HtmlAgilityPack.HtmlDocument();
document.Load(new MemoryStream(File.ReadAllBytes(openFileDialog1.FileName)), UnicodeEncoding.UTF8);
corpus = ExtractViewableTextCleaned(document.DocumentNode).Replace(">", "").Replace(" ", "").Replace(""", "");
//corpus = f1.RemoveTags(corpus);
file.Close();
}
//------------ Start Process ---------
Regex_Patterns.Instance.ClearDigitList();
Regex_Patterns.Instance.ClearEndMarkList();
Regex_Patterns.Instance.ClearSignList();
Regex_Patterns.Instance.ClearSentenceList();
Regex_Patterns.Instance.ClearWordList();
Regex_Patterns.Instance.ClearCurrentSentence();
Regex_Patterns.Instance.Tape = corpus;
Regex_Patterns.Instance.HeadTape = -1;
while (Regex_Patterns.Instance.HeadTape < Regex_Patterns.Instance.Tape.Length - 1)
{
new S0().Compute();
}
DataTable mytable = new DataTable();
mytable.Columns.Add("sentence", typeof(string));
DataRow row;
sentencesOfCorpus = Regex_Patterns.Instance.GetSentenceList;
foreach (string sent in Regex_Patterns.Instance.GetSentenceList)
{
row = mytable.NewRow();
row[0] = sent;
mytable.Rows.Add(row);
}
DataSet ds = new DataSet();
ds.Tables.Add(mytable);
dataGridView1.DataSource = ds.Tables[0].AsDataView();
btnProcessTrigram.Enabled = true;
btnProcessBigram.Enabled = true;
btnProcessUnigram.Enabled = true;
}
private void button3_Click_1(object sender, EventArgs e)
{
GRAMCLASS gc = new GRAMCLASS();
if (MemoryClass.Instance.isEmptyTrigramList())
{
gc.TriGram2(MemoryClass.Instance.Get_Sentences_From_OldCorpus(), true);
}
string currentSentence = string.Empty;
DataTable mytable = new DataTable();
mytable.Columns.Add("Sentence", typeof(string));
mytable.Columns.Add("Probability", typeof(double));
DataRow row;
double SumProbability = 0;
for (int rowIndex = 0; rowIndex < (dataGridView1.Rows.Count - 1); rowIndex++)
{
currentSentence = dataGridView1[0, rowIndex].Value.ToString();
ArrayList SS = gc.Seperator_Phase4(currentSentence, false);
double PDF = 1;
StructData_TriGram dataTrigram = new StructData_TriGram();
for (int i = 2; i < SS.Count; i++)
{
dataTrigram.CurrentToken = ((string)SS[i]).ToLower();
dataTrigram.PrevToken = ((string)SS[i - 1]).ToLower();
dataTrigram.PrevPrevToken = ((string)SS[i - 2]).ToLower();
if (MemoryClass.Instance.GetListTrigram.Exists(s => s.CurrentToken == dataTrigram.CurrentToken && s.PrevToken == dataTrigram.PrevToken && s.PrevPrevToken == dataTrigram.PrevPrevToken))
{
StructData_TriGram trigrm = MemoryClass.Instance.GetListTrigram[gc.ListTrigram.FindIndex(s => s.CurrentToken == dataTrigram.CurrentToken &&
s.PrevToken == dataTrigram.PrevToken && s.PrevPrevToken == dataTrigram.PrevPrevToken)];
int root = (trigrm.Probability != 1) ? trigrm.Probability.ToString().Remove(0, 2).Length : 1;
PDF *= Math.Pow(trigrm.Probability, (double)1 / SS.Count);
}
else
{
PDF = 0;
}
}
row = mytable.NewRow();
row[0] = currentSentence;
row[1] = (PDF); // *((double)currentSentence.Length / Max_Index);
SumProbability += (PDF); // *((double)currentSentence.Length / Max_Index);
mytable.Rows.Add(row);
}
DataSet ds = new DataSet();
ds.Tables.Add(mytable);
dataGridView1.DataSource = ds.Tables[0].AsDataView();
lblAverageResult.Text = ((double)SumProbability / Regex_Patterns.Instance.GetSentenceList.Count).ToString();
}
private void btnOpenFile_Click(object sender, EventArgs e)
{
GRAMCLASS gc = new GRAMCLASS();
string corpus = string.Empty;
Fill_MemoryClass();
openFileDialog1.Filter = "HtmFile (*.htm)|*.htm|Htmlfile (*.html)|*.html|MhtFile(*.mht)|*.mht";
DialogResult dr = openFileDialog1.ShowDialog();
if (dr == DialogResult.OK)
{
StreamReader file = new StreamReader(openFileDialog1.FileName);
corpus = file.ReadToEnd();
Form1 f1 = new Form1();
corpus = f1.RemoveTags(corpus);
file.Close();
}
//------------ Start Process ---------
Regex_Patterns.Instance.ClearDigitList();
Regex_Patterns.Instance.ClearEndMarkList();
Regex_Patterns.Instance.ClearSignList();
Regex_Patterns.Instance.ClearSentenceList();
Regex_Patterns.Instance.ClearWordList();
Regex_Patterns.Instance.ClearCurrentSentence();
Regex_Patterns.Instance.Tape = corpus;
Regex_Patterns.Instance.HeadTape = -1;
while (Regex_Patterns.Instance.HeadTape < Regex_Patterns.Instance.Tape.Length - 1)
{
new S0().Compute();
}
DataTable mytable = new DataTable();
mytable.Columns.Add("sentence", typeof(string));
DataRow row;
sentencesOfCorpus = Regex_Patterns.Instance.GetSentenceList;
foreach (string sent in Regex_Patterns.Instance.GetSentenceList)
{
row = mytable.NewRow();
row[0] = sent;
mytable.Rows.Add(row);
}
DataSet ds = new DataSet();
ds.Tables.Add(mytable);
dataGridView1.DataSource = ds.Tables[0].AsDataView();
btnProcessTrigram.Enabled = true;
btnProcessBigram.Enabled = true;
btnProcessUnigram.Enabled = true;
}
private void button4_Click_1(object sender, EventArgs e)
{
GRAMCLASS gramclass = new GRAMCLASS();
DataTable table = new DataTable();
table.Columns.Add("sentence", typeof(string));
table.Columns.Add("Probability", typeof(double));
DataRow row;
double SumProbability = 0;
int CountOfTotalTokens = MemoryClass.Instance.Get_CountOfTotalTokens();
string sentence = string.Empty;
//foreach (string sentence in Regex_Patterns.Instance.GetSentenceList)
for (int index_row = 0; index_row < dataGridView1.Rows.Count - 1; index_row++)
{
sentence = dataGridView1.Rows[index_row].Cells[0].Value.ToString();
ArrayList al = gramclass.Seperator_Phase4(sentence, false);
double PDF = 1;
foreach (string token in al)
{
int Count = MemoryClass.Instance.Get_CountOfSpecialToken_Unigram(token);
if (Count != -1)
{
PDF *= (double)MemoryClass.Instance.Get_CountOfSpecialToken_Unigram(token) / CountOfTotalTokens;
}
else
{
PDF = 0;
break;
}
}
row = table.NewRow();
row[0] = sentence;
row[1] = PDF;
SumProbability += (PDF);
table.Rows.Add(row);
}
DataSet ds = new DataSet();
ds.Tables.Add(table);
dataGridView1.DataSource = ds.Tables[0].AsDataView();
lblAverageResult.Text = ((double)SumProbability / Regex_Patterns.Instance.GetSentenceList.Count).ToString();
}
private string Select_Token_From_GridView(out ArrayList al, out int token_index, out int row_index)
{
GRAMCLASS gc = new GRAMCLASS();
//------- create row_index randomly --------------------------------------
row_index = Create_Random_index(dataGridView1.Rows.Count - 1, false);
//------ select row randomly and seperate sentence within row -------
al = new ArrayList();
al.AddRange(gc.Seperator_ChangeCorpus(dataGridView1.Rows[row_index].Cells[0].Value.ToString()));
//------ select a token randomly from previous line --------
token_index = Create_Random_index(al.Count - 1, true);
string cur_token = al[token_index].ToString();
if (Regex_Patterns.Instance.GetMarkType(cur_token) != TypeMark.Word && Regex_Patterns.Instance.GetMarkType(cur_token) != TypeMark.Digit)
{
Select_Token_From_GridView(out al, out token_index, out row_index);
}
return(cur_token);
}
private void Switch_tokens_Corpus()
{
GRAMCLASS gc = new GRAMCLASS();
Random rnd_row_index = new Random();
Random rnd_token_index = new Random();
//---------- first select row and token randomly --------
ArrayList al;
int token_index_1;
int row_index_1;
string cur_token = Select_Token_From_GridView(out al, out token_index_1, out row_index_1);
//---------- second select row and token randomly --------
int token_index_2;
int row_index_2;
ArrayList al_new;
string new_token = Select_Token_From_GridView(out al_new, out token_index_2, out row_index_2);
//------ switch tokens --------
al.RemoveAt(token_index_1); al.Insert(token_index_1, new_token);
al_new.Remove(token_index_2); al_new.Insert(token_index_2, cur_token);
//----- fill sentence from al list --------
string current_sentence = dataGridView1.Rows[row_index_1].Cells[0].Value.ToString();
string new_sentence1 = string.Empty;
string new_sentence2 = string.Empty;
for (int al_i = 0; al_i < al.Count; al_i++)
{
new_sentence1 += " " + al[al_i].ToString();
}
for (int al_new_i = 0; al_new_i < al_new.Count; al_new_i++)
{
new_sentence2 += " " + al_new[al_new_i].ToString();
}
dataGridView1.Rows[row_index_1].Cells[0].Value = new_sentence1;
dataGridView1.Rows[row_index_2].Cells[0].Value = new_sentence2;
dataGridView1.Rows[row_index_1].DefaultCellStyle.BackColor = Color.Red;
dataGridView1.Rows[row_index_2].DefaultCellStyle.BackColor = Color.Red;
}
private void Show_Phase4_TO_GridView(Form f, DataSet ds, bool isTrigram)
{
DataTable table1 = new DataTable();
table1.Columns.Add("Setntence", typeof(string));
table1.Columns.Add("Probability", typeof(double));
DataRow r;
DataGridView dg2 = ((DataGridView)f.Controls[f.Controls.IndexOfKey("grid_sentences")]);
GRAMCLASS gl = new GRAMCLASS();
List <StructData_BiGram> bigram = new List <StructData_BiGram>();//= gl.BiGram(Regex_Patterns.Instance.Tape);
List <StructData_TriGram> trigram = new List <StructData_TriGram>();
string sentences = string.Empty;
foreach (string sentence in Regex_Patterns.Instance.GetSentenceList)
{
sentences += sentence;
}
if (isTrigram)
{
trigram = gl.TriGram(sentences.ToLower());
}
else
{
bigram = gl.BiGram(sentences.ToLower());
}
//bool isFirst = true;
double Avg = 0;
foreach (DataGridViewRow row in dg2.Rows)
{
if (row.Cells[0].Value != null)
{
string sentence = string.Empty;
sentence = row.Cells[0].Value.ToString();
ArrayList new_SeperatedSentence = gl.Seperator_Phase4(sentence, isTrigram);// BiGram
r = table1.NewRow();
r[0] = row.Cells[0].Value.ToString();
double currentProbability;
if (isTrigram)
{
currentProbability = gl.Verify_Sentence(new_SeperatedSentence, trigram, isTrigram);
}
else
{
currentProbability = gl.Verify_Sentence(new_SeperatedSentence, bigram, isTrigram);
}
r[1] = currentProbability;
Avg += currentProbability;
table1.Rows.Add(r);
}
}
ds.Tables.Add(table1);
DataView dv = new DataView();
dv = ds.Tables[1].AsDataView();
dv.Sort = "Probability ASC";
dataGridView1.DataSource = dv;
lblAverageResult.Text = ((double)Avg / (dataGridView1.Rows.Count - 1)).ToString();
}
private void BindDataToDataGridView(bool isTriGram, bool isCorpus)
{
string sentences = string.Empty;
DataSet ds = new DataSet();
DataTable mytable = new DataTable();
mytable.Columns.Add("Token", typeof(string));
mytable.Columns.Add("PrevToken", typeof(string));
if (isTriGram)
{
mytable.Columns.Add("PrevPrevToken", typeof(string));
}
mytable.Columns.Add("Probility", typeof(double));
mytable.Columns.Add("Count", typeof(int));
DataRow r;
double Avg = 0;
if (isCorpus)
{
foreach (string sent in Regex_Patterns.Instance.GetSentenceList)
{
sentences += sent;
}
//sentences += Regex_Patterns.Instance.Tape;
if (isTriGram)
{
List <StructData_TriGram> l = new GRAMCLASS().TriGram(sentences);
foreach (StructData_TriGram SD in l)
{
r = mytable.NewRow();
r[0] = SD.CurrentToken;
r[1] = SD.PrevToken;
r[2] = SD.PrevPrevToken;
r[3] = SD.Probability;
r[4] = SD.Count;
mytable.Rows.Add(r);
Avg += SD.Probability;
}
}
else
{
List <StructData_BiGram> l = new GRAMCLASS().BiGram(sentences);
foreach (StructData_BiGram SD in l)
{
r = mytable.NewRow();
r[0] = SD.CurrentToken;
r[1] = SD.PrevToken;
r[2] = SD.Probability;
r[3] = SD.Count;
mytable.Rows.Add(r);
Avg += SD.Probability;
}
}
}
else
{
foreach (string sentence in Regex_Patterns.Instance.GetSentenceList)
{
sentences = sentence;
if (isTriGram)
{
List <StructData_TriGram> l = new GRAMCLASS().TriGram(sentences);
foreach (StructData_TriGram SD in l)
{
r = mytable.NewRow();
r[0] = SD.CurrentToken;
r[1] = SD.PrevToken;
r[2] = SD.PrevPrevToken;
r[3] = SD.Probability;
r[4] = SD.Count;
mytable.Rows.Add(r);
Avg += SD.Probability;
}
}
else
{
List <StructData_BiGram> l = new GRAMCLASS().BiGram(sentences);
foreach (StructData_BiGram SD in l)
{
r = mytable.NewRow();
r[0] = SD.CurrentToken;
r[1] = SD.PrevToken;
r[2] = SD.Probability;
r[3] = SD.Count;
mytable.Rows.Add(r);
Avg += SD.Probability;
}
}
}
}
ds.Tables.Add(mytable);
dataGridView1.DataSource = ds.Tables[0].DefaultView;
lblAverageResult.Text = ((double)Avg / (dataGridView1.Rows.Count - 1)).ToString();
}
