Please enable JavaScript.

Coggle requires JavaScript to display documents.

Chapter 6 (Exemplary Techniques (Nearest neighbor methods (Issues with…

- - - - Taking a majority vote of a target
    - - Important to classify data neighbors based on how similar they are and then weight them appropriately
    - - Using averages to predict values
    - - Dimensionality and Domain Knowledge
      - Computational Efficiency
      - Intelligibility
  - - - This is done by finding natural subgroups within a data set
    - - Organizes by clusters, with some clusters inside of others
    - - k-means clustering
        
        “means” are the centroids, represented by the arithmetic means (averages) of the values along each dimension for the instances in the cluster
        
        'k" is simply number of clusters one would like to analise
- - - - (sqrt((d1,A-d1,B)^2+(d1A-d2B)^2+(dnA-dnB)^2))