Social Network Based User Profiling

Predicting user's preferences based on their and their network's social feed from Facebook & Twitter.

Project Synopsis

A silicon valley start up wanted to develop an analytics platform that could gather intelligence about users' preferences and social behaviour pertaining to their products and offerings. This data could then be used for targeted selling of products and services to the users.

The developed platform was expected to have the following features

  • Provide user segmentation analysis for the application's users on the fly
  • Collect users' data on social network in a non-intrusive manner
  • Provide ability to plug rules to extract different analytics from the system without major application changes

Rare Mile Solution

Rare Mile worked with the client's team to collect publicly available users' data from Facebook and Twitter and apply text analysis rules on top of the data to produce analytics. The solution also used Natural Language Processing (NLP) capabilities to infer likes and dislikes from the data. The system was also integrated with two external websites to enrich the product feed before running analysis on it.

Rare Mile proposed and implemented a solution with the following features:

  • Natural language based data processing
  • Non intrusive data collection and enrichment for analysis
  • Close to real time user segmentation

Project Highlights

  • Designed and implemented the application in 10 weeks
  • Employed iterative development to absorb client's feedback regularly during the application development
  • Used a research oriented approach for trying different candidate solution approaches before deciding the final solution

About The Project

This project involved extracting and predicting user's preferences based on what he or his network is talking about on Facebook and Twitter

Technologies Used

Java Based Backend
Natural Language Processing (NLP)
Open Source Libraries
Python

Client Details