TF-IDF计算四

最新推荐文章于 2024-01-09 13:06:45 发布

原创最新推荐文章于 2024-01-09 13:06:45 发布 · 949 阅读

0 ·

CC 4.0 BY-SA版权

文章标签：

#hashmap #class

本文介绍了一种基于TF-IDF算法的文本特征提取方法，并详细展示了如何通过具体步骤实现TF-IDF算法。包括构建文件列表、计算词频(TF)、逆文档频率(IDF)以及最终的TF-IDF得分。

最后, 开始计算TF-IDF

TF i - IDF i,j = TF i,j * IDF i

代码:

class TFIDF
{
ArrayList<HashMap<String,Double>> TFIDFMainFileList = new ArrayList<HashMap<String,Double>>();
ArrayList<HashMap<String,Double>> TFMainFileList = new ArrayList<HashMap<String,Double>>();
ArrayList<ArrayList<String>> MainFileList = new ArrayList<ArrayList<String>>();
HashMap<String,Double> IDFMainFileList = new HashMap<String, Double>();

public TFIDF(ArrayList<ArrayList<String>> mfl, HashMap<String,Double> idfm, ArrayList<HashMap<String,Double>> tfmfl)
{
  MainFileList = mfl;
  IDFMainFileList = idfm;
  TFMainFileList = tfmfl;
}

public ArrayList<HashMap<String,Double>> PrintTFIDF()
{
  for(int i=0; i<MainFileList.size(); i++)
  {
   ArrayList<String> SubFileList = MainFileList.get(i);
   HashMap<String,Double> tfFile = TFMainFileList.get(i);

   HashMap<String,Double> GetTFIDF = new HashMap<String, Double>();
   ArrayList<String> Index = new ArrayList<String>();

   //Take hashmap level from ArrayList<HashMap<String,Double>>
   for(int j=0; j<SubFileList.size(); j++)
   {

    Index.add(SubFileList.get(j));

    double tf = tfFile.get(SubFileList.get(j));
    double idf = IDFMainFileList.get(SubFileList.get(j));
    double tfidf = tf * idf;

    GetTFIDF.put(SubFileList.get(j),tfidf);

   }
   TFIDFMainFileList.add(GetTFIDF);
   //==========================================================
  }
  return TFIDFMainFileList;
}
}