Symbolic generators for complex networks
Symbolic Generators for Complex Networks
NOTE: If you are looking for the original Java version of Synthetic, you can find it here: https://github.com/telmomenezes/synthetic-old/ . This Python library is under active development, please excuse the current lack of documentation. We will make it available soon.
Synthetic is a machine learning tool that can be used to discover plausible generators for complex networks. Generators are simple computer programs that control the growth of a network from the bottom-up, in a similar fashion to the processes believed to underlie the emergence of many different types of networks, be them biological, social or technological.
Generators are useful as network growth models, both for their potentially explanatory and predictive powers. In a way, this tool automates the scientific method. It creates and refines hypothesis, and tests them against real data, in a process that leads to increasingly plausible models.
Programs are represented in a very simple language that is suitable both for humans and the machine learning process. The machine learning algorithm used is Genetic Programming, belonging to the family of Evolutionary Algorithms, which are inspired by Darwinian evolution. Programs are subject to random variations, much like the genetic material in biological entities. The programs with the highest quality survive, leading to an ongoing refinement of the growth model.
For a more complete explanation see this article: http://www.nature.com/srep/2014/140905/srep06284/full/srep06284.html
Menezes, T. and Roth, C., Symbolic regression of generative network models, Scientific Reports 4 (2014) http://www.nature.com/srep/2014/140905/srep06284/full/srep06284.html
Release history Release notifications | RSS feed
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Hashes for synthetic-0.0.1-py3-none-any.whl