[Question] Could we use recommender systems to figure out human values?

[Olga Babeeva]

Would it make sense to try to figure out human values with recommender systems? Why or why not? How could this be done?