[TRUE / FALSE] K-means clustering algorithms can find clusters of arbitrary shape.

[TRUE / FALSE] The silhouette coefficient is a method to determine the natural number of clusters for partitioning algorithms.

[TRUE / FALSE] DBSCAN clustering algorithm optimize an objective function.

[TRUE / FALSE] A common weakness of association rule mining is that it produces too many frequent itemsets.

[TRUE / FALSE] Lift corrects for high confidence in rule X-> Y when item X is bought regularly by customers.

[TRUE / FALSE] A low support of X will increase the confidence of X -> Y.

[TRUE / FALSE] When we have X -> Y and Y -> X then support of X is equal to support of Y.

[TRUE / FALSE] In general, agglomerative clustering is slower than quadratic.

[TRUE / FALSE] The best centroid for minimizing the SSE of a cluster is the mean of the points in the cluster.

[TRUE / FALSE] Apriori principle indicates if an itemset is infrequent, then all of it subsets must also be infrequent.

[TRUE / FALSE] Sparsification step in graph-based Chameleon algorithm can significantly reduce the effects of noise and outliers.

[TRUE / FALSE] K Nearest Neighbor is considered as a non-parametric method.

[TRUE / FALSE] Artificial neural network is guaranteed to be successfully trained using gradient descent for global optimal.

[TRUE / FALSE] Boosting method produces an ensemble of classifiers through random sampling the training data set with replacement.