public static string GetKeyWordsSplitBySpace(string keywords, PanGuTokenizer ktTokenizer)
{
StringBuilder result = new StringBuilder();
ICollection <WordInfo> words = ktTokenizer.SegmentToWordInfos(keywords);
foreach (WordInfo word in words)
{
if (word == null)
{
continue;
}
result.AppendFormat("{0}^{1}.0 ", word.Word, (int)Math.Pow(3, word.Rank));
}
return(result.ToString().Trim());
}
/// <summary>
/// 盘古分词
/// </summary>
/// <param name="keywords">分词关键字</param>
/// <returns>分词后的字符</returns>
/// <remarks>2013-3-8 杨浩 创建</remarks>
public static string Participle(string keywords)
{
StringBuilder result = new StringBuilder();
PanGuTokenizer ktTokenizer = new PanGuTokenizer();
ICollection <PanGu.WordInfo> words = ktTokenizer.SegmentToWordInfos(keywords);
if (words.Count != 0)
{
_keyword = words.Max().Word;//记录用户查询关键字
}
foreach (var word in words.Where(word => word != null))
{
result.AppendFormat("{0}^{1}.0 ", word.Word, (int)Math.Pow(3, word.Rank));
}
return(result.ToString().Trim());
}
private static string GetKeyWordsSplitBySpace(string keywords, PanGuTokenizer ktTokenizer)
{
StringBuilder result = new StringBuilder();
/*执行分词操作 一个关键字可以拆分为多个次和单个字*/
ICollection <WordInfo> words = ktTokenizer.SegmentToWordInfos(keywords);
foreach (WordInfo word in words)
{
if (word == null)
{
continue;
}
result.AppendFormat("{0}/", word.Word);
}
return(result.ToString().Trim());
}
/// <summary>
/// 字符串分词
/// </summary>
/// <param name="keywords">字符串</param>
/// <param name="ktTokenizer">分词</param>
/// <returns></returns>
public string GetKeyWordsSplitFilter(string keywords, PanGuTokenizer ktTokenizer)
{
// 拆分字符串
List <string> list = new List <string>();
ICollection <WordInfo> words = ktTokenizer.SegmentToWordInfos(keywords);
foreach (WordInfo word in words)
{
if (word == null)
{
continue;
}
else if (word.Word.Length > 1 && !list.Contains(word.Word))
{
list.Add(word.Word);
}
}
return(string.Join(" ", list.ToArray()));
}
public void ContainAnalyzerTest()
{
LuceneEngine engine = new LuceneEngine();
engine.Init();
//string text = "三国演义";
//string text = "a|f|g|i|m";
string text = "NBA常规赛-快船vs凯尔特人";
TextReader tr = new StringReader(text);
PanGuTokenizer ct = new PanGuTokenizer(tr);
int end = 0;
Lucene.Net.Analysis.Token t;
string ss = string.Empty;
while (end < text.Length)
{
t = ct.Next();
end = t.EndOffset();
ss = ss + t.TermText() + "/ ";
}
string xxx = ss;
}
public void SpellTest()
{
LuceneEngine engine = new LuceneEngine();
engine.Init();
var firsts = SpellUtils.GetCnSegment("NBA常规赛-快船vs凯尔特人");
var spells = SpellUtils.GetSpellSegment("战重警和");
var segment = new Segment();
var collection = segment.DoSegment("国际足球100509K联赛釜山-大田");
var list = new List <string>();
foreach (WordInfo word in collection)
{
if (word == null)
{
continue;
}
list.Add(word.Word);
}
//var list = SpellUtils.GetSpellSegment("战警");
string sss = "san国yan义";
string ssss = Synacast.LuceneNetSearcher.Searcher.Searcher.SegmentKeyWord(sss);
string text = "重庆";
TextReader tr = new StringReader(text);
PanGuTokenizer ct = new PanGuTokenizer(tr);
int end = 0;
Lucene.Net.Analysis.Token t;
string ss = string.Empty;
while (end < text.Length)
{
t = ct.Next();
end = t.EndOffset();
ss = ss + t.TermText() + "/ ";
}
}
