Classification methods that are based on vector spaces are not necessarily the best option for problem sets that are represented using exclusively nominal feature sets. Which of the above claims do you agree with the most, and which do you disagree with the most? Which classification approach would you use if you were doing ML classification on nominal data? Please offer comprehensive responses to the two questions presented in the preceding sentence.

Classification methods that are based on vector spaces are not necessarily the best option for problem sets that are represented using exclusively nominal feature sets. Which of the above claims do you agree with the most, and which do you disagree with the most? Which classification approach would you use if you were doing ML classification on nominal data? Please offer comprehensive responses to the two questions presented in the preceding sentence.