ASAP Please, answer only, no need to explain

Which one is NOT correct k-means clustering? As we increase k in k-means clustering, the sum of squared errors will decrease. A good clustering solution maximizes within-cluster variation. OK-means clustering can be used for market segmentation. O The number of clusters needs to be specified to initiate a k-means clustering algorithm. To run a 4-cluster k-means clustering, you selected the following 4 initial cluster centroids: (7, 5), (6, 3), (2, 5), (4, 3). A point (0,4) will be assigned to The third cluster centroid The first cluster centroid The fourth cluster centroid The second cluster centroid LD Select the best number of clusters using the elbow method in k-means clustering (x: number of clusters, y: sum of squared distance to the cluster centroid of each sample) 3250 2000 1750 Squared Error 1500 1250 1600