Clustering - The Data Ensemble MCQ's
1. What is a preferred distance measure while dealing with sets ?
View Answer
Jaccard
2. Each point is a cluster in itself. We then combine the two nearest clusters into one. What type of clustering does this represent ?
View Answer
Agglomerative
3. Which learning is the method of finding structure in the data without labels.
View Answer
Unsupervised
4. ____________ of a set of points is defined using a distance measure .
View Answer
Similarity
5. Members of the same cluster are far away / distant from each other .
View Answer
False
6. Unsupervised learning focuses on understanding the data and its underlying pattern.
View Answer
True
7. ___________ of two points is the average of the two points in Eucledian Space.
View Answer
Centroid
8. A centroid is a valid point in a non-Eucledian space .
View Answer
False
9. What is the overall complexity of the the Agglomerative Hierarchical Clustering ?
View Answer
O(N^3)
10. ___________ measures the goodness of a cluster.
View Answer
Cohesion
11. ___________ is the data point that is closest to the other point in the cluster.
View Answer
Clusteroid
12. The ______ is a visual representation of how the data points are merged to form clusters.
View Answer
Dendogram
13. _____________ is when points don't move between clusters and centroids stabilize.
View Answer
Convergence
14. ___________ is a way of finding the k value for k means clustering.
View Answer
Cross Validation
15. The number of rounds for convergence in k means clustering can be lage.
View Answer
True
16. Sampling is one technique to pick the initial k points in K Means Clustering.
View Answer
True
17. K Means algorithm assumes Eucledian Space/Distance.
View Answer
True
18. What is the R Function to divide a dataset into k clusters ?
View Answer
kclusters()
19. What is the R function to apply hierarchical clustering to a matrix of distance objects ?
View Answer
hclust()
No comments:
Post a Comment