I would build a scheudled job that imports the taxonomy file from wherever it's stored then loads it in a processes it. The file can then process it using the category API to create/modify categories where needed and then yoou can programatically update pages using the content repository to tag them (if you have some sort of identifier between makes and the import data).
If you need to store any extra data on categories, I'd suggest using https://github.com/Geta/EpiCategories as it allow modelling of category information the same way you do other IContent items.
We have a taxonomy managed by our organisation, and the taxonomy is stored in a file. The taxonomy is a tree consisting of categories (for example School category has subcategories Primary School and Secondary School, and Primary School has subcategories State Primary School and Private Primary School and so on). We would like Episerver to import the taxonomy, and use the taxonomy to categorise pages in Episerver for search etc. If the taxonomy changes in future, Episerver can reimport the taxonomy. Any suggestion is welcome. Thanks!