You might try recursive partitioning for a directly interpretable model. Extreme gradient boosting is state-of-the-art but will be more steps to harder to interpret, like here, with the neural net.
Sent from my iPhone On Jul 16, 2022, at 12:46 PM, Jochen Fromm <[email protected]> wrote: It is the first time I am using the SHAP package and I am not completely sure how to interpret the result https://shap.readthedocs.io/en/latest/index.html I would say humiliation is definitely an essential factor. In Germany before the emergence of Nazism there was for instance a feeling of national humiliation after the loss of WW I and the Treaty of Versailles. Robert Paxton also mentions it in his book about the anatomy of fascism. -J. -------- Original message -------- From: Marcus Daniels <[email protected]> Date: 7/15/22 11:57 PM (GMT+01:00) To: The Friday Morning Applied Complexity Coffee Group <[email protected]> Subject: Re: [FRIAM] Modeling Fascism Humiliation is not predictive of fascism, am I reading that right? From: Friam <[email protected]> On Behalf Of Steve Smith Sent: Friday, July 15, 2022 1:29 PM To: [email protected] Subject: Re: [FRIAM] Modeling Fascism for those of us trying to suss out the implications of this, I offer: https://en.wikipedia.org/wiki/Shapley_value I'm not clear on a number of things Jochen: 1. The Shapley values implies an iterated/cooperative model? 2. I see how the "training set" from post WWI Europe might be used in this context. 3. How do your assumptions about Trump-era America fit in? * Is the assumption that post WWI (hidden) conditions in Europe are similar enough to contemporary Global conditions? I'm probably missing something (a lot)? - Steve On 7/15/22 2:06 PM, Jochen Fromm wrote: Today after work I've tried to built a small and simple machine learning model to predict fascism, based on 6 or 7 fundamental features. Using Shapley values we can see which feature contributes the most to the outbreak of fascism. If I enter values which fit to the US under Trump the model indeed predicts a form of authoritarianism. No big data or deep learning, just a small neural network based on scikit-learn (no Tensorflow, Keras, or Pytorch). Juypter notebook here: https://nbviewer.org/github/JochenFromm/JupyterNotebooks/blob/master/ModelingFascism.ipynb -J. -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. . FRIAM Applied Complexity Group listserv Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoom https://bit.ly/virtualfriam to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com FRIAM-COMIC http://friam-comic.blogspot.com/ archives: 5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/ 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/ -. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. . FRIAM Applied Complexity Group listserv Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoom https://bit.ly/virtualfriam to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com FRIAM-COMIC http://friam-comic.blogspot.com/ archives: 5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/ 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
-. --- - / ...- .- .-.. .. -.. / -- --- .-. ... . / -.-. --- -.. . FRIAM Applied Complexity Group listserv Fridays 9a-12p Friday St. Johns Cafe / Thursdays 9a-12p Zoom https://bit.ly/virtualfriam to (un)subscribe http://redfish.com/mailman/listinfo/friam_redfish.com FRIAM-COMIC http://friam-comic.blogspot.com/ archives: 5/2017 thru present https://redfish.com/pipermail/friam_redfish.com/ 1/2003 thru 6/2021 http://friam.383.s1.nabble.com/
