Algorithms For Dummies

Considering the need for remediation

After you find problems with your dataset, you need to remediate it so that the 

dataset works properly with the algorithms you use. For example, when working 

with conflicting data types, you must change the data types of each data source so 

that they match and then create the single data source used with the algorithm. 

Most of this remediation, although time consuming, is straightforward. You sim-

ply need to ensure that you understand the data before making changes, which 

means being able to see the content in the context of what you plan to do with it. 

However, you need to consider what to do in two special cases: data duplication 

and missing data. The following sections show how to deal with these issues.