In: Computer Science
What domain expertise specific information is collected and organized during the problem definition step of a data mining spiral?
Defining the problem is the first step in the data mining spiral model. In this step, the information that is collected includes defining the scope of the problem, analyzing business requirements, defining specific objectives for the data mining project, and defining the matrices by which the model will be evaluated. The information that is collected can be translated into questions as follows:
1. What you want exactly? What types of relationships are you trying to find?
2. The problem that you are trying to solve reflects the policies of the business?
3. What attributes you want to try to predict?
4. What kind of data and information you have in each column? If there is more than one table then how are the tables related? To make the data usable, do you need to perform any aggregation or processing?
5. Does the data represent the processes of business accurately? How data is distributes?