back

supreme court analysis slowdownOCT 23 2023

fuzzy string matching and struggles

i just got rapidfuzz to work in matching differing supreme court case names in two datasets! one has a column of facts for each court case, and one has all the ideology score calculations i made for each year. the idea is that: given the facts of the case, will the models improve? right now they sort of suck :( now i'm working on dividing my datasets into smaller ones to see if training my models on smaller datasets improves them. after that, i'll start merging the two datasets so court case facts can be used in training.