1. You are beginning a information discovery project in a domain you know well. You know the data sources and the limitations of what the data represents. Naturally, the data youve selected and collected includes many attributes. Among these attributes are (a) record identifiers (e.g., patient id, test id), and (b) attributes known to have noisy data (these data are collected from equipment with some inherent noise)
   1. What do you do with the record identifier(s) when performing a data-mining project, if anything? When is this (the thing you said you would do) done? (i.e., Before data mining? after data mining? mixed in? explain. If you dont do anything, explain why not.