/cf-viz

Collaborative Filtering and Force-Directed Graph Visualization of Congress

Primary LanguageHTMLMIT LicenseMIT

Graph Network Visualization + Recommendation Engine

Collaborative Filtering and Force-Directed Graph Visualization

The goal of this project is to explore concepts of non-negative matrix factorization and cosine similarity, as seen in some recommender systems so that I can take data scraped from Congress and recommend the best person to cosponsor a bill with for a congressman. This project is written using python and d3.js

To view the visualization, I suggest the following code: python -m SimpleHTTPServer 8000 You have to run this simple server because of the Cross origin request error (the files are being server locally and hence no http calls are being made)

Here's what you should see:

![cfviz] (https://dl.dropbox.com/s/7ww0cd5at5j9j78/Screen%20Shot%202015-10-23%20at%205.44.25%20AM.png?dl=0)