Analytics Visualizer

Implemented quite a few more algorithms

One of the challenging parts about this sprint was that I realized a bit too late that just knowing all of the scores doesn’t help because the user still has to filter through everything. This is why, I had to change many of the functions to return the top n scores.

There were also many variations on algorithms for computational efficiency. I implemented multiple versions for now and will decide which ones (or all!) do I want to keep in the final version.

Now, we go to the drawing board and figure out how to make my limited JS and Streamlit knowledge work with me as I attempt to create a frontend.

Initial Setup!

To preface, as a graph analytics intern and competitive programmer, a huge part of my life is working with graph algorithms.

And what’s the hardest part about working with them? Visualizing when to use them.

Enter Analytics Visualizer: Take the algorithms you know, and see how they work and when to use them.

At the moment, I unfortunately don’t know how I’m going to get the visualization part to work (I’m clearly a bit more of a backend person), but on the bright side, I got the basic outline of the algorithms part!

We have a class for the graph object, and then another method that converts it to something usable in networkx. Currently, I’m working on making methods for all of the algorithms. So far, I have bfs, dfs, degree, dijkstra, and triadic closure. My goal for today is to finish this part!