Case Studies

Intent Mining on Social Media


Background & Challenges

  • The challenge was the mine and maintain the huge amounts of data on a daily basis
  • Filter out actual intent – rather than indicated intent of donation of members


Our Approach

  • Created a web based crawler to extract data from Facebook and Twitter
  • Used Natural language Processing and Machine Learning techniques to extract posts indicating intent of donation
  • Used Hadoop and Map Reduce to manage and process huge amount of data generated


Deliverables

  • API to extract the potential donor data on real time basis
  • Demographic Profiling of potential donors
  • Donors interested in cross donations and up-donations
  • Most popular causes and campaigns based on member engagement