求助：Term-Doc Matrix 到 Term-Term Matrix

YangJ

谢谢。但是想用自己的Matrix，具体定义是这样的，不知道哪里错了。

<br />
library(rJava)<br />
library(tm)<br />
library(rmmseg4j)<br />
dt <- DirSource("C:/Users/welcome/Desktop/zf", encoding ="GBK")<br />
ovid<- Corpus(dt)<br />
CMetaData(ovid)<br />
DMetaData(ovid)<br />
reuters <- tm_map(ovid, stripWhitespace)<br />
zj <- c("maxword")<br />
re <- 0<br />
for (i in 1:dt$Length) {<br />
re[[i]]<- mmseg4j(PlainTextDocument(reuters)[[i]],zj)<br />
}<br />
reuters <- Corpus(VectorSource(re))<br />
dtm <- TermDocumentMatrix(reuters, control=list(wordLengths=c(1,Inf)))<br />

YangJ

<br />
> save (dtm, file ="dtm.Rdata")<br />
> load(file="dtm.Rdata")<br />
> dtm[dtm>=1] <- 1<br />
Error in dtm[dtm >= 1] <- 1 : invalid subscript type 'list'<br />
> dtm <- dtm %*% t(dtm)<br />
Error in dtm %*% t(dtm) :<br />
  requires numeric/complex matrix/vector arguments<br />

nan.xiao

由于某种原因，dtm 的数据类型不对。气氛上看应该是个矩阵的感觉。

YangJ

我想也应该是的了，不过资质太浅，不知如何修改，请赐教。

nan.xiao

无数据无真相。。。

YangJ

这么个咚咚：

<br />
> dtm<br />
$i<br />
  [1]  4  6 15 35 39 41 42 51 55 61 62 70 74 79 82 84 85 88 91 92  8 14 19 21<br />
 [25] 22 24 25 28 31 32 35 38 40 43 50 59 60 66 69 71 73 76  5 18 20 29 30 53<br />
 [49] 56 57 65 67 68 82 87 90 95  1  2  5  9 11 12 17 18 23 33 45 46 47 48 49<br />
 [73] 52 54 58 64 69 75 77 81 82 83 86 89  3  5  7 10 13 16 18 26 27 34 36 37<br />
 [97] 44 57 63 69 72 78 80 93 94</p>
<p>$j<br />
  [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2<br />
 [37] 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4<br />
 [73] 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5</p>
<p>$v<br />
  [1] 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1<br />
 [37] 1 1 3 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1<br />
 [73] 1 1 1 1 2 1 1 1 1 1 1 1 1 2 1 1 1 1 2 1 1 5 1 1 1 1 1 1 1 1 1 1 1</p>
<p>$nrow<br />
[1] 95</p>
<p>$ncol<br />
[1] 5</p>
<p>$dimnames<br />
$dimnames$Terms<br />
 [1] "t"    "ta"   "白天" "拜拜" "不"   "不必" "不是" "不孝" "厕所" "承认"<br />
[11] "冲去" "床上" "次数" "打"   "大家" "但我" "倒"   "的"   "底下" "地铁"<br />
[21] "点点" "电话" "掉"   "都不" "都在" "对了" "多的" "躲"   "方法" "飞"<br />
[31] "胳膊" "跟"   "捍卫" "呼哧" "就"   "就是" "开心" "哭"   "啦"   "懒觉"<br />
[41] "离开" "没"   "每次" "每个" "每天" "闷"   "梦到" "尿"   "暖"   "强大"<br />
[51] "去死" "权利" "让"   "上好" "什么" "使我" "是"   "誓死" "谁"   "睡"<br />
[61] "死"   "所以" "太"   "躺"   "讨厌" "特别" "天上" "条"   "我"   "我的"<br />
[71] "我妈" "我是" "我一" "我有" "无门" "想"   "小时" "笑"   "要的" "也挺"<br />
[81] "夜里" "一"   "一壶" "抑郁" "原因" "再"   "在"   "在意" "找"   "整"<br />
[91] "症"   "重要" "自己" "自然" "坐"  </p>
<p>$dimnames$Docs<br />
[1] "1" "2" "3" "4" "5"</p>
<p>attr(,"class")<br />
[1] "TermDocumentMatrix"    "simple_triplet_matrix"<br />
attr(,"Weighting")<br />
[1] "term frequency" "tf"<br />

nan.xiao

dtm = as.matrix(dtm) 一下试试。

YangJ

不能强迫输入'double'，不知何意？

<br />
> dtm[dtm>=1] <- 1<br />
Error in dtm[dtm >= 1] <- 1 :<br />
  (list) object cannot be coerced to type 'double'<br />
> dtm <- dtm %*% t(dtm)<br />
Error in dtm %*% t(dtm) :<br />
  requires numeric/complex matrix/vector arguments<br />

nan.xiao

<br />
require(tm)<br />
data(crude)<br />
tdm = TermDocumentMatrix(crude)</p>
<p>tdm$i = c(4, 6, 15, 35, 39, 41, 42, 51, 55, 61, 62, 70, 74, 79, 82, 84, 85, 88, 91, 92, 8, 14, 19, 21, 22, 24, 25, 28, 31, 32, 35, 38, 40, 43, 50, 59, 60, 66, 69, 71, 73, 76, 5, 18, 20, 29, 30, 53, 56, 57, 65, 67, 68, 82, 87, 90, 95, 1, 2, 5, 9, 11, 12, 17, 18, 23, 33, 45, 46, 47, 48, 49, 52, 54, 58, 64, 69, 75, 77, 81, 82, 83, 86, 89, 3, 5, 7, 10, 13, 16, 18, 26, 27, 34, 36, 37, 44, 57, 63, 69, 72, 78, 80, 93, 94)<br />
tdm$j = as.integer(c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5))<br />
tdm$v = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 3, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 1, 1, 1, 1, 2, 1, 1, 1, 1, 2, 1, 1, 5, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1)<br />
tdm$nrow = 95L<br />
tdm$ncol = 5L<br />
tdm$dimnames$Terms = c("t", "ta", "白天", "拜拜", "不", "不必", "不是", "不孝", "厕所", "承认", "冲去", "床上", "次数", "打", "大家", "但我", "倒", "的", "底下", "地铁", "点点", "电话", "掉", "都不", "都在", "对了", "多的", "躲", "方法", "飞", "胳膊", "跟", "捍卫", "呼哧", "就", "就是", "开心", "哭", "啦", "懒觉", "离开", "没", "每次", "每个", "每天", "闷", "梦到", "尿", "暖", "强大", "去死", "权利", "让", "上好", "什么", "使我", "是", "誓死", "谁", "睡", "死", "所以", "太", "躺", "讨厌", "特别", "天上", "条", "我", "我的", "我妈", "我是", "我一", "我有", "无门", "想", "小时", "笑", "要的", "也挺", "夜里", "一", "一壶", "抑郁", "原因", "再", "在", "在意", "找", "整", "症", "重要", "自己", "自然", "坐")<br />
tdm$dimnames$Docs = c("1", "2", "3", "4", "5")</p>
<p>dtm = as.matrix(tdm)<br />
dtm[dtm >= 1] = 1<br />
dtm = dtm %*% t(dtm)<br />

YangJ

谢谢啊！你是把内容一个个手动输入进去的吗？终于实现word的社交网了，再谢一个！就是不知道这些字的输入能否自动完成？

YangJ

回复第11楼的 nan.xiao：弄好了，再次感谢。其实你一开始说as.matrix是对的。只是之前还要加一个as.data.frame，呵呵。