C# ICTCLAS 中文分词 源码

由吕震宇根据Free版ICTCLAS，在CLR平台下CSharp语言的实现

using System;using System.Collections.Generic;using System.Text;using SharpICTCLAS;public class WordSegmentSample{ private int nKind = 1; //在NShortPath方法中用来决定初步切分时分成几种结果 private WordSegment wordSegment; //======================================================= // 构造函数，在没有指明nKind的情况下，nKind 取 1 //======================================================= public WordSegmentSample(string dictPath) : this(dictPath, 1) { } //======================================================= // 构造函数 //======================================================= public WordSegmentSample(string dictPath, int nKind) { this.nKind = nKind; this.wordSegment = new WordSegment(); //wordSegment.PersonRecognition = false; //wordSegment.PlaceRecognition = false; //wordSegment.TransPersonRecognition = false; //---------- 订阅分词过程中的事件 ---------- wordSegment.OnSegmentEvent = new SegmentEventHandler(this.OnSegmentEventHandler); wordSegment.InitWordSegment(dictPath); } //======================================================= // 开始分词 //======================================================= public List<WordResult[]> Segment(string sentence) { return wordSegment.Segment(sentence, nKind); } //======================================================= // 输出分词过程中每一步的中间结果 //======================================================= private void OnSegmentEventHandler(object sender, SegmentEventArgs e) { switch (e.Stage) { case SegmentStage.BeginSegment: Console.WriteLine("\r\n==== 原始句子：\r\n"); Console.WriteLine(e.Info "\r\n"); break; case SegmentStage.AtomSegment: Console.WriteLine("\r\n==== 原子切分：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.GenSegGraph: Console.WriteLine("\r\n==== 生成 segGraph：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.GenBiSegGraph: Console.WriteLine("\r\n==== 生成 biSegGraph：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.NShortPath: Console.WriteLine("\r\n==== NShortPath 初步切分的到的 N 个结果：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.BeforeOptimize: Console.WriteLine("\r\n==== 经过数字、日期合并等策略处理后的 N 个结果：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.OptimumSegment: Console.WriteLine("\r\n==== 将 N 个结果归并入OptimumSegment：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.PersonAndPlaceRecognition: Console.WriteLine("\r\n==== 加入对姓名、翻译人名以及地名的识别：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.BiOptimumSegment: Console.WriteLine("\r\n==== 对加入对姓名、地名的OptimumSegment生成BiOptimumSegment：\r\n"); Console.WriteLine(e.Info); break; case SegmentStage.FinishSegment: Console.WriteLine("\r\n==== 最终识别结果：\r\n"); Console.WriteLine(e.Info); break; } }}