向量空间模型文档相似度计算实现(c).docVIP

using System; using System.Collections.Generic; using System.Linq; using System.Text; using System.Text.RegularExpressions; namespace Felomeng.VSMSimilarity { class SVMModle { /// summary /// 降维词表 /// /summary private Liststring reducingKeys = new Liststring(); /// summary /// 构造函数：使用降维表 /// /summary /// param name=reducingKeys降维词表/param public SVMModle(Liststring reducingKeys) { this.reducingKeys = reducingKeys; } /// summary /// 构造函数：不使用降维表 /// /summary public SVMModle() { } /// summary /// 相似度计算 /// /summary /// param name=text1文档１（分好词的，分词符为非汉字字符）/param /// param name=text2文档２（分好词的，分词符为非汉字字符）/param /// returns两篇文章的相似度/returns public double Similarity(string text1, string text2) { double similarity = 0.0, numerator = 0.0, denominator1 = 0.0, denominator2 = 0.0; int temp1, temp2; Dictionarystring, int dictionary1 = GetDictionary(text1); Dictionarystring, int dictionary2 = GetDictionary(text2); if ((dictionary1.Count 1) || (dictionary2.Count 1))//如果任一篇文章中不含有汉字 { return 0.0; } Dictionarystring, int.KeyCollection keys1 = dictionary1.Keys; foreach (string key in keys1) { dictionary1.TryGetValue(key, out temp1); if (!dictionary2.TryGetValue(key, out temp2)) { temp2 = 0; } dictionary2.Remove(key); numerator += temp1 * temp2; denominator1 += temp1 * temp1; denominator2 += temp2 * temp2; } Dictionarystring, int.KeyCollection keys2 = dictio