Clustering Flashcards
1
Q
Clustering
A
Searches for natural grouping
2
Q
Distances
A
Lnorm Eucldean Manhattan Linf_norm Jaccard Cosine Distance Edit = |x| + |y| - 2|LCS| (longest common subsequence) Hamming
3
Q
Scikit - Learn Encoders
A
Label Encoder 0,1,……n-1
OneHotEncoder
OrdinaEncoder
4
Q
Requisites for Clustering
A
Scalability Different types of attributes Handle dynamic data Arbitrary Shape Deal Noise and outliers High dimensionality
