The project's objective is to develop an algorithmic strategy that might forecast if a customer would select one of the 10 financial packages your business is providing. You have access to the 5,000-entry database of prior clients maintained by the business. The database includes 30 measured customer characteristics for each customer.

You've chosen to utilize the k-nearest neighbor (k-nn) classifier to finish the task.

What steps would you take to create the K-Nn classifier? How would you go about testing and training the model? How are you going to choose the best value for k? What are the k-nn algorithm's shortcomings? Give at least two examples and elaborate on each one.