Please enable JavaScript.

Coggle requires JavaScript to display documents.

Chapter 6: Similarity, Neighbors, and Clusters (Nearest -Neighbor…

- - - - From here, we run regression
- - - - lack of an explicit, interpretal model
      - could cause problems in complex data mining
    - - numeric values may have drastically different values
      - inaccurate distance calculations between variables
      - irrelevant attributes- such a the feature selection
    - - expensive/ not efficient to store large databases
- - - - scales, kinds
    - - The Manhattan- sums the different dimensions
      - The Jacquard
        
        measures text distances
      - Cosine Distance
        
        represents the norm/ length of each figure
    - - calculates equations to predict distance
- - - - represent each by the center of their cluster
      - uses the k-mean theorem, retrieves desired amount of clusters
        
        an algorithm