Skip to content

Latest commit

 

History

History
16 lines (13 loc) · 1.17 KB

README.md

File metadata and controls

16 lines (13 loc) · 1.17 KB
Raw

Text_and_Sentiment_Analytics

This repository contains the code file, data-set and a insights report on using text analytics on Amazon.com employee reviews, a training covered in Intelligent Machines and is available on: https://www.youtube.com/watch?v=oVIl2-K0O5Y&t=3221s

Background

Text data can be very rich and informative, and can help companies gain very interesting insights. This tutorial, code and report gives us an idea of what we can understand from the text data, and how it can be valuable for businesses. The context is that of employee reviews about a workplace, both pros and cons.

Content

  1. Work Frequency analysis
  2. Work Frequency Clouds, Commonality Clouds, Comparison Clouds
  3. Understanding which terms are connected with other terms through association networks and association scores
  4. Understanding the overall sentiment of the organization
  5. Document wise sentiment and emotions
  6. Topic Modeling - getting an idea of what themes people are talking about

Potential Future Work

These analyses can be applied to understand policy perception or customer perception and after effects of marketing and policy campaigns. Any other ideas are welcome here!