Find the maximum and minimum values for each behavior: This is needed in order to normalize the dataset.

To do this, complete the findMostSimilarCustomer() method located in the SimilarCustomer.java file. The findMostSimilarCustomer() will take the arguments double[][] customerBehaviorMatrix, and int currentCustomerIndex. The method will return a matrix called mostSimilarCustomerIndex.

1

The dataset needs to be normalized so that each behavior/property is measured on the same scale. Normalization for each entry can be done using the following formula:

^{Formula for normalization}

where x_ij is the entry of the i^th customer with j^th behavior and min_j and max_j are the minimum and maximum values of the j^th behavior.

3

Calculate similarity between customers. This can be done using the basic Euclidean distance formula.

Xjj - min; -,Vij mai max; min; Xjj - min; -,Vij mai max; min