This phase includes both understandings of the bank business and the bank who needs to understand how machine learning works in order to avoid misunderstanding in case of an unexpected result. Before starting it is important that the bank understands that not always there is a solution and even if one is found, there will be a probability that the result for a data point is not corrected. There will be a level of accuracy...
This phase includes both understandings of the bank business and the bank who needs to understand how machine learning works in order to avoid misunderstanding in case of an unexpected result.
Before starting it is important that the bank understands that not always there is a solution and even if one is found, there will be a probability that the result for a data point is not corrected. There will be a level of accuracy that the model can receive and the bank needs to decide if it can be accepted. To do that it is necessary to understand the business of the bank and the data the bank provided. It is helpful to know if the customer can provide more data, in case they do not cover all the possible situation or if they are unbalanced (e.g., 80% of them are Unemployed, the result will be that there are more possibilities for an Unemployed person to do not repay the loan).
