Question

作为arules算法的数据准备的一部分，我有以下格式的数据。

  public class GraphicalBoard extends JComponent { 

     private int xSize = 50; 
     private int ySize = 50; 
     private Board board;
     private int size;

      public GraphicalBoard() {
       } 

      public void setBoard(Board board){
         this.board = board;
     } 

      public void paintComponent(Graphics g) { 
        super.paintComponent(); 

        if(board == null) {
          throw new RuntimeException("Board not set") ;
       } 

         String colors[] = {"BLUE","GREEN","YELLOW","RED","BLACK","ORANGE","PINK"}; 
         Graphics2D g2 = (Graphics2D) g; 
         int xCoord = 0; 
         int yCoord = 0; 
         int colorNum = 0; 
         int size = board.getSize() ;
        String colorOfSquare = ""; 

        for (int count = 0; count < size; count++) { 

            for (int counter = 0; counter < size; counter++) { 
               colorNum = board.getBoard(count, counter) ; 
               colorOfSquare = colors[colorNum]; 
               g2.setColor(colorOfSquare); 
               Rectangle square = new Rectangle(xCoord,yCoord,xSize,ySize); 
               xCoord += 50; 
             } 
           yCoord += 50;
         } 
}

我需要交易文件

userid p1 p2 p3
1      1  0  0 
2      1 1   0
3      0  1  1
4      0  1  0

你能帮我解决这个问题。

感谢。

Answer 1

一种选择是在逻辑which上使用matrix并提取row.names

row.names(which(t(df1[-1])==1, arr.ind = TRUE))
#[1] "p1" "p1" "p2" "p2" "p3" "p2"

Answer 2

您需要做的就是将数据转换为logical。

 library("arules")

 dat <- data.frame(
   userid = 1:4, 
   p1 = c(1,1,0,0), 
   p2 = c(0,1,1,1), 
   p3 = c(0,0,1,0)
 )

 dat2 <- dat[,-1]
 for(i in 1:ncol(dat2)) dat2[[i]] <- as.logical(dat2[[i]])
 dat2

      p1    p2    p3
 1  TRUE FALSE FALSE
 2  TRUE  TRUE FALSE
 3 FALSE  TRUE  TRUE
 4 FALSE  TRUE FALSE

 tran <- as(dat2, "transactions")
 inspect(tran)

     items   transactionID
 [1] {p1}    1            
 [2] {p1,p2} 2            
 [3] {p2,p3} 3            
 [4] {p2}    4

arules算法的数据准备

2 个答案: