Third Step!

HomeThird Step!

So, this is the final step. I built a website for users that wants to search for small graphlets up to 4 nodes in large 300+ networks with different domains from social science to biology.

First things first, we need to install a framework for communicating between backend and frontend. I used Flask with Python and mostly HTML, all on Ubuntu 20.04 LTS. Flask is easy to work with even if you do not have any background in web development and confused with the server and client connections.

Installing Flask on your Ubuntu

First, make sure you have the most recent version of Python. If so, then open the terminal and type:

$ pip install Flask

for running the project, I did a simple command line as follows:

export FLASK_APP=flaskr
export FLASK_ENV=development
flask run
Goals
  • Design and implement a user-friendly system for graphlet detection.
  • A user can choose datasets (real/artificial) and can choose graphlets for a given dataset.
  • Skip the process of installing the system and dependencies.
  • Using the best and fastest Graphlet counter engine for large/small networks (PGD).
  • Being able to choose a specific k-vertex graphlet to count.

The result

The user just need to choose a specific graphlet to count. Then hit the “Search” button and finally see the result in red.

Technical parts
  • Using Flask which is a micro and light web framework for simple web applications like this.
    • Developed by Armin Ronacher, written in Python
    • Supports extensions for object-relational mappers, upload handling, and others
    • No database abstraction layer
    • Not require particular tools or libraries (microframework)
  • The search engine is Parallel Parameterized Graphlet Decomposition (PGD).
    • Developed by scientists at Purdue University and Texas A&M University
    • Algorithms: A fast, efficient, and parallel graphlet counting algorithm
    • Scalability: Achieves on average 460x runtime improvement o
    • Effectiveness: Over 300+ large-scale networks from a variety of domains
    • Applications: Finding unique patterns in graphs and graph similarity and classification.