R：topicmodels，2个相似的文档，代码与其中一个工作，不与另一个

Question

当我运行我的topicmodel代码时，发生了一个非常奇怪的错误。 基本上我有一个包含用户注释的.csv文件。我想创建一个dtm，每个注释都是一个文档。我采取了8K评论的样本，并使用以下代码：

> #LOAD LIBRARYS 
> 
> library(tm) 
> library(SnowballC) 
> library(stringr) 
> library(tictoc) 
> tic() 
> 
> #SET FILE LOCATION 
> file_loc <- "C:/Users/Andreas/Desktop/first8k.csv" 
> 
> #LOAD DOCUMENTS 
> Database <- read.csv(file_loc, header = FALSE) 
> require(tm) 
> 
> #PROCEED 
> Database <- Corpus(DataframeSource(Database)) 
> 
> Database <-tm_map(Database,content_transformer(tolower)) 
> 
> 
> Database <- tm_map(Database, removePunctuation) 
> Database <- tm_map(Database, removeNumbers) 
> Database <- tm_map(Database, removeWords, stopwords("english")) 
> Database <- tm_map(Database, stripWhitespace) 
> 
> 
> myStopwords <- c("some", "individual", "stop","words") 
> Database <- tm_map(Database, removeWords, myStopwords) 
> 
> Database <- tm_map(Database,stemDocument) 
> 
> 
> dtm <- DocumentTermMatrix(Database,   control=list(minDocFreq=2,minWordLength=2)) 
> 
> row_total = apply(dtm, 1, sum) 
> dtm.new = dtm[row_total>0,] 
> 
> removeSparseTerms(dtm, .99) 
> 
>>Outcome:DocumentTermMatrix (documents: 12753, terms: 194) 
>Non-/sparse entries: 66261/2407821 
>Sparsity   : 97% 
>Maximal term length: 11 
>Weighting   : term frequency (tf) 
> 
> #TOPICMODELLING 
> 
> library(topicmodels) 
> 
> 
> 
> burnin <- 100 
> iter <- 500 
> thin <- 100 
> seed <-list(200,5,500,3700,1666) 
> nstart <- 5 
> best <- TRUE 
> 
> 
> k <- 12 
> 
> 
> ldaOut <-LDA(dtm.new,k, method="Gibbs", control=list(nstart=nstart, seed = seed, best=best, burnin = burnin, iter = iter, thin=thin)) 
> 

所以这一个工程就好了。如果我拿的8K的意见，也csv文件，格式相同另一个样品等则出现下列错误：

> library(tm) 
> library(SnowballC) 
> library(stringr) 
> library(tictoc) 
> tic() 
> 
> #SET FILE LOCATION 
> file_loc <- "C:/Users/Andreas/Desktop/try8k.csv" 
> 
> #LOAD DOCUMENTS 
> Database <- read.csv(file_loc, header = FALSE) 
> require(tm) 
> 
> #PROCEED 
> Database <- Corpus(DataframeSource(Database)) 
> 
> Database <-tm_map(Database,content_transformer(tolower)) 
> 
> 
> Database <- tm_map(Database, removePunctuation) 
> Database <- tm_map(Database, removeNumbers) 
> Database <- tm_map(Database, removeWords, stopwords("english")) 
> Database <- tm_map(Database, stripWhitespace) 
> 
> 
> myStopwords <- c("some", "individual", "stop","words") 
> Database <- tm_map(Database, removeWords, myStopwords) 
> 
> Database <- tm_map(Database,stemDocument) 
> 
> dtm <- DocumentTermMatrix(Database,control=list(minDocFreq=2,minWordLength=2)) 
> 
> row_total = apply(dtm, 1, sum) 
> dtm.new = dtm[row_total>0,] 
> 
> removeSparseTerms(dtm, .99) 
> 
>>Outcome:DocumentTermMatrix (documents: 9875, terms: 0) 
Non-/sparse entries: 0/0 
Sparsity   : 100% 
Maximal term length: 0 
Weighting   : term frequency (tf) 
> 
> #TOPICMODELLING 
> 
> library(topicmodels) 
> 
> 
> 
> burnin <- 100 
> iter <- 500 
> thin <- 100 
> seed <-list(200,5,500,3700,1666) 
> nstart <- 5 
> best <- TRUE 
> 
> 
> k <- 12 
> 
> 
> ldaOut <-LDA(dtm.new,k, method="Gibbs", control=list(nstart=nstart, seed = seed, best=best, burnin = burnin, iter = iter, thin=thin)) 

>Fehler in obj[[i]][[which.max(sapply(obj[[i]], logLik))]] : 
>attempt to select less than one element in get1index 

我想与DTM自己是不是沃金，因为它说，有9875个文档，但根本没有条件。但我完全不知道为什么代码适用于一个样本，但不适用于另一个样本。请告诉我，如果我在代码上做了错误的事，或者发现了其他错误。

提前致谢！

Answer 1

terms = 0这就是为什么你有可能