有什么方法可以在相似性函数中组合多个距离度量？

Question

I need to find a way to code a similarity function between two vector(data instances), let's name them and . These data instances have categorical features as well as quantities. Thus, I'd like to find a way to combine, let's say Hamming Distance and Euclidean Distance as to use it in my association problem.

There are merging types for k-NN, such as voting etc., still association problem cannot be solved by voting approach.

Answer 1

混合数据类型时，可以通过Gower距离查找距离。