>library(tm) //使用默認安裝的R平臺是不帶tm package的，必須要到http://www.r-project.org/網站下載package. 值得注意的是：tm package很多函數也要依賴於其它的一些package,所以在這個網站，應該把rJava,Snowball,zoo,XML,slam,Rz,Rweka,matlab這些win32 package一併下載，並解壓到默認的library中去。
>vignette("tm") //會打開一個tm.pdf的英文文件，講述tm package的使用及相關函數
1、Data-import：
> txt <- system.file("texts", "txt", package = "tm") //是爲將目錄C:\Program Files\R\R-2.15.1\library\tm\texts\txt 記入txt變量
> (ovid <- Corpus(DirSource(txt),readerControl = list(language = "lat"))) //即將txt目錄下的5個文件Corpus到Ovid去，language = "lat"表示the directory txt containing Latin (lat) texts
此外，VectorSource is quite useful, as it can create a corpus from character vectors, e.g.:
> docs <- c("This is a text.", "This another one.")
> Corpus(VectorSource(docs)) //A corpus with 2 text documents
在本部分中，我們Finally create a corpus for some Reuters documents as example for later use
> reut21578 <- system.file("texts", "crude", package = "tm")
> reuters <- Corpus(DirSource(reut21578),readerControl = list(reader = readReut21578XML)) // 在這一部分中，將目錄C:\Program Files\R\R-2.15.1\library\tm\texts\crude下的20個XML文件Corpus成reuters，要用到XML package(前面已經下載了).
> inspect(ovid[1:2]) //會出現以下的顯示，當然identical(ovid[[2]], ovid[["ovid_2.txt"]])==true，所以inspet(ovid["ovid_1.txt","ovid[ovid_2.txt]"])效果一樣：
2、Transmation：
> reuters <- tm_map(reuters, as.PlainTextDocument) //This can be done by converting the documents to plain text documents.即去除標籤
> reuters <- tm_map(reuters, stripWhitespace) //去除空格
> reuters <- tm_map(reuters, tolower) //將內容轉換成小寫
> reuters <- tm_map(reuters, removeWords, stopwords("english")) // remove stopwords
注：在這裏需要注意的是，如果使用中文分詞法，由於詞之間無有像英文一樣的空隔，好在有Java已經解決了這樣的問題，我們只需要在R-console里加載rJava與rmmseg4j兩個工具包即可。如
>mmseg4j("中國人民從此站起來了")
[1] 中國 人民 從此 站 起來
3、Filters:
> query <- "id == '237' & heading == 'INDONESIA SEEN AT CROSSROADS OVER ECONOMIC CHANGE'" //query其實是一個字符串，設定了一些文件的條件，如
//id==237, 標題爲：indonesia seen at c.........
> tm_filter(reuters, FUN = sFilter, query) // A corpus with 1 text document,這個從數據中就可以看得出來。
4、Meta data management
> DublinCore(crude[[1]], "Creator") <- "Ano Nymous" //本來第一個XML文件中是不帶作者的，此語句可以改變一些屬性的值，類比其它。
> meta(crude[[1]]) //顯示第一個文件的元素信息數據得到下圖
> meta(crude, tag = "test", type = "corpus") <- "test meta"
> meta(crude, type = "corpus") 改變元素後顯示如下
5、Creating Term-Document Matrices
> dtm <- DocumentTermMatrix(reuters)
> inspect(dtm[1:5, 100:105]) //顯示如下：
A document-term matrix (5 documents, 6 terms)
Non-/sparse entries: 1/29
Sparsity : 97%
Maximal term length: 10
Weighting : term frequency (tf)
Terms
Docs abdul-aziz ability able abroad, abu accept
127 0 0 0 0 0 0
144 0 2 0 0 0 0
191 0 0 0 0 0 0
194 0 0 0 0 0 0
211 0 0 0 0 0 0
6、對Term-document矩陣的進一步操作舉例
> findFreqTerms(dtm, 5) //nd those terms that occur at least 5 times in these 20 files 顯示如下：
[1] "15.8" "accord" "agency" "ali"
[5] "analysts" "arab" "arabia" "barrel."
[9] "barrels" "bpd" "commitment" "crude"
[13] "daily" "dlrs" "economic" "emergency"
[17] "energy" "exchange" "exports" "feb"
[21] "futures" "government" "gulf" "help"
[25] "hold" "international" "january" "kuwait"
[29] "march" "market"
> findAssocs(dtm, "opec", 0.8) // Find associations (i.e., terms which correlate) with at least 0:8 correlation for the term opec
opec prices. 15.8
1.00 0.81 0.80
如果需要考察多個文檔中特有詞彙的出現頻率，可以手工生成字典，並將它作爲生成矩陣的參數
> d <- Dictionary(c("prices", "crude", "oil")))
> inspect(DocumentTermMatrix(reuters, list(dictionary = d)))
因爲生成的term-document矩陣dtm是一個稀疏矩陣，再進行降維處理，之後轉爲標準數據框格式
> dtm2 <- removeSparseTerms(dtm, sparse=0.95) //parse值越少，最後保留的term數量就越少
> data <- as.data.frame(inspect(dtm2)) //最後將term-document矩陣生成數據框就可以進行聚類等操作了見下部分
7、 再之後就可以利用R語言中任何工具加以研究了，下面用層次聚類試試看
> data.scale <- scale(data)
> d <- dist(data.scale, method = "euclidean")
> fit <- hclust(d, method="ward")
>plot(fit) //圖形見下：
