Design for a machine-learning artificial intelligence framework

Question

This is a community wiki that aims to provide a good design for a machine learning/artificial intelligence framework (ML/AI framework).

Please contribute to the design of a language-agnostic framework which would allow multiple ML/AI algorithms to be plugged into a single framework which:

• runs the algorithms with a user-specified data set.

• facilitates learning, qualification, and classification.

• allows users to easily plug in new algorithms.

• can aggregate or create an ensemble of the existing algorithms.

• can save/load the progress of the algorithm (i.e. save the network and weights of a neural network, save the tree of a decision tree, etc.).

What is a good design for this sort of ML/AI framework?

Answer 1

Possibly one can start by looking at the design of existing open-source ML/AI frameworks. To name a few: Weka, RapidMiner, KNIME, Orange.

If you wish to learn more about Machine Learning and Artificial Intelligence
then visit this Artificial Intelligence Master Training Course.