thaohp113
commented
1 month ago Our main focus is to make machine learning more inclusive to other people, especially non-tech people. Therefore, I think the USP is making everything as intuitive as possible. Even though there are many tutorials / tools for visualizing, we think that they are not accessible for the broad audience, since they require certain knowledge about ML and even some prior set up.
Regarding our idea, we’ll add a few visualizations to each algorithm and help users interact with the visualization. For example, we will attempt visualizing data at a higher dimension (3D), and have visualizations of both train/test data, or allowing users to see side-by-side comparisons of how different algorithms perform on the same dataset - could be extremely informative.
Clear project idea/ contribution and working plan
A few things to note:
On the internet, can you find similar tools that visualize how basic ML algorithms work? If you can then how your idea is a unique one and can add value to the field of data visualization? When i was in school, i rmb that we also have to build a shiny app to visualize a kmeans clustering problem, so I am just wondering how your idea is different from that (maybe a diverse set of algorithms user can choose? Or you can add a few visualizations for each algorithms to help user analyze/ understand how to analyze the output of each algorithm?)