Slide background


Applying ML for Text Classification in CRM

Based on our experience at Kreato on applying ML techniques, we see that Text analytics or automatic text classification can be utilized on CRM for the below use cases.

  • Segment customer data
  • Perform sentiment analysis
  • Perform intent analysis
  • Suggest contextual responses
  • Route incoming leads

Basically the text analytics concepts will help to categorize the text data into some ore-defined categories and tag them accordingly to be used for the above use cases.

Data cleansing and text preprocessing have to be performed before applying the text classification. It normally involves replacing special characters and punctuation marks with spaces, normalizing case, removing duplicate characters, removing user-defined or built-in stop-words and word stemming.

For binary classification cases such as sentiment analysis, where you may want to tag an incoming email from customer as either positive or negative, Two-Class Logistic Regression or Two-Class Support Vector Machine algorithms can be tried.

For multiple classification cases such as intent analysis, where you may want to tag an incoming email from customer as a query, support request or requesting for a demo schedule, Multi-class Logistic Regression and Multi-class Decision Forest algorithms can be tried.

To improve accuracy, we could better validate the result yielded with at least two algorithms for the use cases that we would like to try.