A library to help my girlfriend create mindmaps from keywords for her design homework.
Designers often need to create mind-maps to understand how they can make a product give a feeling (for example, Apple products give a feeling of professionalism, sleekness). A mind-map is just a tree with the trigger word as the root. The children of a word node are the words that come to mind on hearing the word.
This project provides a basic UI (a React app) and lets the user automatically create a mind map using some initial trigger words. There's not much direct machine learning involved here - the code will just fetch a combination of synonyms/related words of the parent word and use those as the children, thus continuing the process. The words are fetched from the Datamuse api
The tree is rendered using vis.js, a graph/network rendering library.
- Add option to specify depth of mind-map.
- Use multi-threading to populate branches concurrently.
- Figure out if there's a way to do this without using Datamuse.
- Improve layout (I'm bad at CSS!)
- Add integer input to let the user specify mindmap depth.
- See if its possible to delete nodes from the graph, through the UI.