Machine Learning Catalogue

Methods and Use Cases

This is a catalogue of machine learning methods. It is intended for use when you wish to obtain a list of methods that would be appropriate for a specific use case or when you wish to find out about a specific technique that is mentioned in an article, program or lecture.

Categories

The Machine Learning Catalogue consists of multiple categories.

Algorithm
A machine learning algorithm describes a procedure to train a machine learning model based on training or input data. It also describes the procedure within this model while the model is evaluated at runtime.
(Generic) Use Case
A generic use case describes a domain unspecific application of machine learning functionality to common business problem.
Functional Building Block
A functional building block describes the function an algorithm or a supporting technique fulfills.
Supporting Technique
Supporting techniques are procedures that can each be used to modulate the behavior of a variety of algorithms. Typically they optimize the quality or the performance of algorithms.
Learning Style
A learning style describes how an algorithm or a supporting technique is trained to create a machine learning model.
Output Data Type
An output data type describes the data structure of the data an algorithm would return to a program that has called it.
Internal Model
The internal model is the structure that an algorithm uses for its calculations.
Input Data Type
An input data type describes the data structure and in some cases also the data types that the algorithm would be fed by a program using it.
Relevance
The relevance describes whether an algorithm or a supporting technique is still used for new machine learning solutions.
Parametricity
The parametricity describes to what extent the user specifies the model to the algorithm and to what extent the algorithm learns the model itself.

Generic Use Cases

Free Poster!

We have grouped the Generic Use Cases for Artificial Intelligence and put them together in a poster. The poster is available free of charge as a download and can be printed in DIN A1 format.

Contributors

The Machine Learning Catalogue is the result of the work of the following experts from msg Applied Technology Research.

Richard Paul Hudson
Richard Paul Hudson

Principal IT Consultant,
msg Applied Technology Research

Carol Gutzeit
Carol Gutzeit

Principal IT Consultant,
msg Applied Technology Research

Philip Seiffert
Philip Seiffert

Senior IT Consultant,
msg Applied Technology Research

Mark Lubkowitz
Mark Lubkowitz

Senior IT Consultant,
msg Applied Technology Research