Clustering Game

What is the goal of cluster analysis?

• Get a meaningful intuition of the structure of the data we’re dealing with.

• Cluster-then-predict where different models will be built for different subgroups if we believe there is a wide variation in the behaviors of different subgroups.

• To assign observations to groups so that observations within each group are similar to one another, and the groups themselves stand apart from one another.

• All of the above

Which of the following algorithm is most sensitive to outliers?

• K-means clustering algorithm

• DBSCAN clustering algorithm

• hierarchical clustering algorithm

• all of the above

Different initializations may lead to different clusters in K-means algorithm:

• True

• False

The choice of k, the number of clusters to partition a set of data points:

• Is a personal choice that shouldn't be discussed in public

• Should always be as large as your computer system can handle

• Depends on why you are clustering the data and could use a field expert opinion.

• Has maximum of 10

K-means is an iterative algorithm, and two of the following steps are repeatedly carried out in its inner-loop. Which two?

• Assign each point to its nearest cluster

• Test on the cross-validation set

• Update the cluster centroids based the current assignment

• Using the elbow method to choose K

…. Is a clustering method that is used in machine learning to separate clusters of high density from clusters of low density

• K-means clustering algorithm

• DBSCAN clustering algorithm

• Hierarchical clustering algorithm

• All of the above

Select the parameters that you have to select in advance when using DBSCAN :

• Inertia

• epsilon

• Delta

• min_samples

What is the definition of a border point in DBSCAN:

• If it is within distance epsilon of a core point but has less than min_samples points within distance epsilon

• If at least min_samples points are within distance epsilon (including itself)

• If it is within distance epsilon of a core point but has more than min_samples points within distance epsilon

• If at least min_samples points are within distance epsilon (not including itself)

DBSCAN can identify points that are not part of any cluster

• True

• False

Which algorithm is most suitable for these data points?

• K-means

• DBSCAN