Want to Stop Rebuilding "Expensive" Parts of your Flow? Explicit Builds are the Answer!READ MORE

IME - Leveraging NLP to Analyze Customer Reviews Collected From SMS and Calls

Name: Mohamed AbdElAziz Khamis Omar

Title: Head of Data Science

Country: Egypt

Organization: IME

IME is a key market player in Data Management solutions. To learn more: www.infme.com

Awards Categories:

  • Data Science for Good
  • Moonshot Pioneer(s)

 

Business Challenge:

Arabic Sentiment Analysis to one of the Biggest banks in KSA.

 

Business Solution:

Using the built-in NLP features in Dataiku as well as Facebook word embeddings "wiki-news-300d-1M-subword.bin".

image44.png

 

image47.png

 

image48.png

image50.png

 

image53.png

 

image55.png

 

image58.png

 

image60.png

 

image62.png

 

image64.png

 

image43.png

 

Business Area: Accounting/Finance

Use Case Stage: Proof of Concept

 

Value Generated:

Ability to analyze customer reviews collected from SMS and Calls with high AUC ROC = 0.881.

image23.png

 

image24.png

 

 

Value Brought by Dataiku:

Built-in Natural Language Processing and preparation.

image26.png

 

unnamed (1).png

 

image30.png

 

Value Type:

  • Reduce cost
  • Reduce risk
  • Save time
Share:
Version history
Last update:
2 weeks ago
Updated by: