Suppose you have a dataset of customer transactions from an online retail store. Each data point represents a customer and contains two features: "Total Amount Spent" (in pounds) and "Total Number of Items Purchased." You want to divide the customers into different groups based on their spending behaviour.Which of the following statements about K-means clustering applied to this dataset is true?Group of answer choicesThe number of clusters (K) is determined by the mean of "Total Amount Spent" and "Total Number of Items Purchased."K-means is sensitive to the initial placement of cluster centres, so it's essential to initialise them randomly.K-means will always produce the same clustering result, regardless of the initial positions of the cluster centres.K-means is not suitable for clustering real-valued data and can only handle categorical features.

The correct statement about K-means clustering applied to this dataset is: "K-means is sensitive to the initial placement of cluster centres, so it's essential to initialise them randomly."

Here's why:

1. The number of clusters (K) in K-means is not determined by the mean of any features. It is a parameter that we need to set before we start the algorithm.

2. K-means clustering does indeed depend on the initial placement of the cluster centres. If they are initialised randomly, we might get different results in different runs of the algorithm.

3. Because of the random initial placement of cluster centres, K-means does not always produce the same clustering result.

4. K-means can handle real-valued data. In fact, it's commonly used with real-valued data. It calculates the mean of the points in a cluster to determine the cluster centre, which wouldn't be possible with categorical features.