0

Hi I have a dataset where a call centre agent types the comments against a client ID. We have to classify these comments into different categories based on common words in them. For e.g. "customer wants refund" or "customer is not happy wants a refud" , these two sentences should be one category as it is relating to refunds. I want to use R for classification of these comments. I tried to use KNN or SVM for the same but the problem is that we donot have sample classification to train the Algorithm. Is it possible to classify these comments into common categories based on commonality.

Any help on this will be really appreciated.

Thanks, AB

user7795628
  • 1
  • 1
  • 1
    So what type of model do you want to run in R exactly? Right now this doesn't seem like a programming question that's appropriate for Stack Overflow. If you need help choosing a statistical model, you should be asking over at [stats.se] or [datascience.se]. – MrFlick Mar 31 '17 at 03:56
  • Have you considered something like [sentiment analysis](http://uc-r.github.io/sentiment_analysis) maybe ? – Aramis7d Mar 31 '17 at 04:41

0 Answers0