Sentiment Mining

Cloud-based Sentiment Analysis Platform to analyze both mainstream and social media coverage of key issues.


Business Problem

A leading university wanted to analyze both mainstream and social media coverage of key political issues . The analysis needed to include the evaluation and mapping of the media landscape from several perspectives and based on large-scale data collection of media stories published on the web and shared on Twitter.



Cloud-based Sentiment Analysis Platform

  • Help users to perform detailed multilingual sentiment analysis of texts from different sources – News RSS Feed and social Media sources.
  • Used NLU / P, machine learning algorithms to extract, identify, & characterize the sentiment content
  • Language detection and Parsing performed through Sentiment Analysis Engine & Machine Learning
  • Majority of AWS NLP component rebuilt


Business ROI

  • Time-Saving : Using Predikly’s NLP tool the report was produced in 4-6 weeks, where in it used to take more than 8 months.
  • Increased Efficiency : Instead of a team of 15 people , the report could be built using a team of 4 people.
  • $s Saving : Resulting in saving of $150 K for each report , university if planning to release 6 such reports this year, resulting $900K in saving every year.
  • Increased Throughput : Ability to produce more reports in a shorter amount of time.


Tools and Technologies

  • Python

  • PostgreSQL

  • Mongo DB

  • AMCharts

  • Python-Flask

  • AngularJs

  • HTML

  • CSS

Analysed Performance






RSS Feeds



To know more about our projects in your domain.

"Very dedicated team and highly motivated to understand the business proposition behind the project. Sticks to deadlines and very positive people inside the team. Pleasure to work with."

. . .


Joinville AB . Sweden

Let's create your story

Share your ideas with us and it’s our responsibility to make it real..!!