Team Members: Miranda Rivera and Omar Sagga

This project build on top of the work done in Project 2. In this project, we provide a webservice that implements a backend server to compute the k-mean on a provided dataset loaded from a mongo database. The frontend side of this webservice provides an interactive website that allows the user to try to input a different values for the number of Uber express pickup points (translates to the number of clusters k), then it displays the corresponding graph that it fetches from the backend server.

Given the heavy computation time of k-means, the backend server can run in trail mode, where it only takes a subset of the whole dataset used. It also implements a caching mechanism so that graphs with the same k value don't need to be re-computed, and served immediately.

First make sure you have Mongo database setup (follow the guidelines in here)

• Run the execute script as follows to populate the mongo database

  cd ..
  python3 execute.py mriver_osagga [-t]

  use the -t for trail mode (faster runtime)
• After populating the database, you can simply run the Flask based webserver as follows:

  cd mriver_osagga/webserver/
  python3 webserver.py

  • Now the webserver will be serving on localhost:5000

• The poster can be found here
• The final report can be found here