"Preprocessing of Comments for Sentiment Analysis"

Team,

I want to do preprocessing of the Comments which I want to use further for Sentiment Analysis.
How I will be able to do Preprocessing (includes conversion of Comment to lower case, remove stopwords, remove punctuations [i,e, :,.# @etc]) from the comments which I had collected in .csv file.

Is the sentiment analysis possible in Rapidminer (i.e. Classify any particular comment into Positive or Negative sentiment)

Thanks,

Amit

Find more posts tagged with

AI Studio

Sentiment Analysis

Accepted answers

All comments

MariusHelf

Hi Amit,

you can do that with the text mining extension. For information and tutorials, please have a look at http://rapid-i.com/content/view/189/212/lang,en/

There are videos which cover your exact problems.

Best regards,
Marius

Labi

Hello,

1. How I can connect Linguistic Tokens/Sentences, I tried this tokenization, but no results?

2. How can I filter the text to show me only (more than ) 2 < total occurrences?

3. I am using Open WordNet Dictionary to read the WordNet folder locally from the directory?

Thank you in advance,
Labi

MariusHelf

Hi Labi!

Labi wrote:

1. How I can connect Linguistic Tokens/Sentences, I tried this tokenization, but no results?

What do you mean be "connect"? What does "no result" mean? As always, it may be a good idea to describe more detailed what you expect and attach your process xml as explained in my signature.

2. How can I filter the text to show me only (more than ) 2 < total occurrences?

Again, please describe more detailed what you want to do.

3. I am using Open WordNet Dictionary to read the WordNet folder locally from the directory?

I don't know what you are doing, but did you try out our wordnet extension? Maybe that already helps? It's available from the marketplace at http://marketplace.rapid-i.com

Sorry for answering your questions with only more questions, but I need more info

Best regards,
Marius