There is also specialization since some services that are relatively inactive in the Bitcoin economy are larger contributors to the economy of BitcoinCash. For instance, any given month Google search contributes between 41% and 49% to the group of economies, and the share of Bitcoin is between 85% and 94% of those flows. But in turn, early supporters of the new coin focused on BCH (for instance Bitcoin.com, operator of a popular wallet supporting BCH, went from contributing 87% in August 2017 to 91% in January 2018 -- although its total contribution to the group was marginal).
Results
Due to its ability to identify and focus on driving variables, Symbolic Regression can build models from data sets that have more variables than records (these are commonly known as fat arrays, and most non-evolutionary machine learning techniques find issues to deal with them) \cite{2016}. Therefore, we apply genetic programming for dimensionality reduction purposes, and to build the predictive models that can provide insight into the shape of the trust data space.
Bitcoin
In total, 532 models were generated, with the majority of those (55.1%) containing at least three variables. The modeling process explores the trade-off between model complexity and model error (1-R^2). This is illustrated in the ParetoFrontLogPlot which displays each of the returned models' quality metrics, complexity, and accuracy. The models denoted by red dots are all optimal in the sense that for a given level of accuracy there is no simpler model or, conversely, for a given level of complexity there is no more accurate model \cite{Kotanchek}. Notably, there are 3 models at an order of magnitude materially better than the rest (error on a scale under 10-15), and one of those is an optimal model.