Skip to content

Question about topic modelling

4 messages · Mehdi Dadkhah, John Kane

#
Hi,
I hope you are doing well!
I have a question about topic modeling. Please consider summarized steps
for making a LDA (Latent Direchlet Allocation) model:
1-importing data
2-making a corpus.
3-pre-processing and cleaning data
4-making term document matrix
5-Apply LDA in topicmodel package.
During mentioned steps, should i convert my data to Tidy format using
unnest_tokens or not?

Many thanks!
With best regards,
#
Hi,
I hope you are doing well!
I read a vignette (
https://cran.r-project.org/web/packages/SentimentAnalysis/vignettes/SentimentAnalysis.html)
about interested package, "SentimentAnalysis". But i faced with a question.
In mentioned  vignette, the sentiment has been applied on a sentence or
multiple sentences separately. Can this package calculate sentiment
direction/score for a long texts?
for example:

# Create a vector of strings
documents <- "Wow, I really like the new light sabers!That book was
excellent.R is a fantastic language.The service in this restaurant was
miserable.This is neither positive or negative."

# Analyze sentiment
sentiment <- analyzeSentiment(documents)

Many thanks!
With best regards,
#
I think your best bet is to ask the author/maintainer, Stefan Feuerriegel
,about this. The reference manual
https://cran.r-project.org/web/packages/SentimentAnalysis/SentimentAnalysis.pdf
gives his email address as <sentiment at sfeuerriegel.com>
On Fri, 8 May 2020 at 08:32, Mehdi Dadkhah <mehdidadkhah91 at gmail.com> wrote:

            

  
    
#
Thank you!
On Fri, May 8, 2020 at 10:04 PM John Kane <jrkrideau at gmail.com> wrote: