ML-eduJS is the simple library for using and understand Machine Learning Algorithms - for now decission tree and learning algorithm for them. I would like to invite you to read the code and understand what's happening under the hood. It works only on server side but maybe in future I will create browser side version.
With library you can:
- implement own decision tree with all of the nodes and leafs
- use predefined algorithm for decision tree learning - CART, ID3, C4.5
- save tree to json file with readable form
- save tree to json with form which can be reused later
- descriptive mode
WARNING: Speed and optimization is not the main goal of the library
To install package using npm:
npm install ml_edu_js --save
After that I recommend to assign some properties:
const MLeduJS = require('ml_edu_js');
const CART = MLeduJS.CART; // the same for other submodules which are used in code
With ML-eduJS you can build, test and use your own decision trees. Library implements basic structure, but also offers learning algorithms and ensemble algorithms.
Available learning algorithms:
- CART
- ID3
- C4.5
Available ensemble algorithms:
- bagging
- random forest
More details you can find in Wiki: