Datasets for Comparative Mythology

Do you have a conversation topic that doesn't seem to fit any of the other conversations? Here is where we discuss ANYTHING about Joseph Campbell, comparative mythology, and more!

Moderators: Clemsy, Martin_Weyers, Cindy B.

Locked
jamesd
Posts: 0
Joined: Sat Oct 31, 2015 3:54 pm

Datasets for Comparative Mythology

Post by jamesd »

Looking to apply Machine Learning to a comparative mythology dataset. I work for Zillow which is widely known for its generating home purchase price estimates based on a plethora of criteria including per capita crime rate, number of rooms, age of building, median value of surrounding homes, etc => https://archive.ics.uci.edu/ml/datasets/Housing. The idea is that if a dataset of symbolic forms attributes could be generated for myths, it should be possible to generate a classifier for each stage of the myth based on the other known attributes. The goal would be that for an individual, one could translate their experience into these symbolic forms, probably through a long survey, then the classifier, trained on the myths of the ages, would then get applied to the new scenario and predict the next stage in that individual's personal myth, or maybe even predict other attributes such as the form of bliss that is inherent in that person's experience.

Not saying that these things are solvable absolutely, but wouldn't it be similar to talking to a wise person, in this framework wisdom would be that wise person's internal predictive matrix that hey have coalesced across their lifetime of experience. I think such a algorithm could be formed from a comparative mythology dataset, selecting the attributes would be paramount, but just looking at the Adventure of a Hero section in a hero with a thousand faces, I see the chapters and subsections as possible valid attribute definitions, if one could get a significant data set of consistent scenarios where each stage has a symbolic form, it shouldn't even require translating them to a numerical system directly, then a classifier could be trained based on the forms and the above goal of using the classifier on a person's individual experience should be possible, at least in my mind it would be potentially a significant predictive framework that would use comparative mythology to speak to the lives of individuals, just need to find/build a dataset, anyone know if anything like that has been compiled or put into a machine readable form i.e. spreadsheet?

Locked