Free and Open Machine Learning Book¶
This book is all about applying machine learning solutions for real practical use cases. This means the core focus is on outlining how to use machine learning in a simple way so you can benefit of this powerful technology.
Machine learning is an exciting and powerful technology. The continuous use and growth of machine learning technology opens new opportunities. This great technology should available to use for everyone. This means that everyone should be able to learn, play and create great applications using machine learning technology.
The key focus of this publication in on Free and Open Machine learning technologies only. This to remove barriers for learning, playing and using machine learning technologies for real practical use cases for everyone.
Of course you can use or switch to a large commercial tech giant cloud company to deploy your machine learning driven application in production. However besides vendor lockin, crucial aspects like safety, privacy and security for machine learning application can only be realized when using fully transparent Free and Open technology.
By using outlining an open architecture, FOSS software and open datasets this publication outlines how machine learning technology can be available and usable for everyone. So without barriers. Since the majority of humans are not a graduated mathematician, we skip deep mathematical concepts of machine learning algorithms in this publication. Books with lots of mathematical background information on how machine learning works are available for more than 70 years. There are plenty excellent free and open publications available if you want to learn in depth the inner working of mathematical algorithms powering nowadays machine learning applications. In the learning resources section of this publication you can find some very good references. All available under a creative commons license (cc-by).
This publication has a core focus on how machine learning can be used. This is done by describing:
- Key machine learning concepts. The focus is on concepts that are needed in order to use solid FOSS machine learning frameworks and datasets when creating a machine learning powered application.
- An open reference architecture for creating and maintaining your machine learning application.
- Important open Solution Building Blocks (FOSS based) which you can use to create your machine learning application as fast as possible. Applying machine learning should be as easy and simple as possible. Only when barriers for using this technology are lowered many more great applications will be developed for the benefit for everyone.
- Key quality aspects for engineering and maintaining your machine learning driven application.
- Important safety, privacy and security aspects to prevent disasters
- Ethical issues (like bias) and how to handle this in a transparent way.
This book is created to give you a head start to use and apply Open Source machine learning technologies to solve problems. Without any strings attached, so the focus is on using open transparent technologies only!
This document is in alfa-stage!! Collaboration is fun, so Help Us by contributing ! There are some chapters currently written and editing work (typos,spelling) is yet to be done! Some more background information of the project can be found in the readme on github.com. And do not forget to join the ROI movement!
Table of Contents¶
- Why open machine learning
- What is machine learning
- ML, AI and NLP: What is what
- Statistics is not machine learning
- The paradigm shift: Creating smart software
- Overview machine learning methods
- What is a machine learning model
- Other common terms used in the ML world
- Machine Learning for Business Problems
- When to use machine learning for business problems?
- Common business use cases
- Example use cases
- Exiting ML business examples
- Business Challenges
- Business capabilities needed
- Business ethics
- ML Reference Architecture
- The machine learning process
- Architecture Building Blocks for ML
- Machine learning architecture metamodel
- Security,Privacy and Safety
- Catalogue of Open ML Software
- Acumos AI
- Apache MXNet
- Apache Spark MLlib
- Cookiecutter Data Science
- Data Science Version Control (DVC)
- NLP Architect
- NNI (Neural Network Intelligence)
- OpenCV: Open Source Computer Vision Library
- TextBlob: Simplified Text Processing
- What-If Tool
- Catalogue of Open NLP Software
- ML Learning resources
- NLP Learning resources