diff options
Diffstat (limited to 'SI/Resource/Fundamentals of Data Mining/Content/Compare and Contrast.md')
| -rw-r--r-- | SI/Resource/Fundamentals of Data Mining/Content/Compare and Contrast.md | 27 |
1 files changed, 27 insertions, 0 deletions
diff --git a/SI/Resource/Fundamentals of Data Mining/Content/Compare and Contrast.md b/SI/Resource/Fundamentals of Data Mining/Content/Compare and Contrast.md new file mode 100644 index 0000000..d72dd09 --- /dev/null +++ b/SI/Resource/Fundamentals of Data Mining/Content/Compare and Contrast.md @@ -0,0 +1,27 @@ +--- +id: Compare and Contrast +aliases: + - clustering algorithms +tags: + - Compare-and-Contrast +--- + +## [[clustering algorithms]] + +- [[K-Means]] vs [[K-Medoids]] + - In _K-means_ algorithm, they choose means as the centroids but in the + _K-medoids_, data points are chosen to be the medoids[^1]. +- [[K-Means]] vs [[K-Medians]] + +| K-Means | K-Medians | +| ---------------------------------------------------------- | --------------------------------------------- | +| The center is not necessarily one of the input data points | Centers will be chosen from data points | +| Not flexible | More flexible | +| Not immune to noise and outliers | More robust to noise and outliers | +| Minimize the sum of squared Euclidian distance | Minimize a sum of pairwise of dissimilarities | + +[^1]: + Medoids are **representative objects of a data set or a cluster within a + data set whose sum of dissimilarities to all the objects in the cluster is + minimal**. Medoids are similar in concept to means or centroids, but medoids are + always restricted to be members of the data set. |