Fuzzy Modeling and Genetic Algorithms for Data Mining and Exploration

6.10 Compatibility Threshold Management

■

187

Thus, using a fuzzy query approach instead of a conventional SQL

query we have found three customer candidates instead of one. Indeed,

the fuzzy query selected customers that matched the intent of our query

rather than the strict algebraic rules of the query. As we can see, as the

number of column attributes increases the brittleness of conventional SQL

queries also rapidly increases. The fuzzy query approach strikes a balance

in the process by including records that to some degree meet the intent of

our query. Selecting the minimum or average result analysis then allows

us to choose records that have some ...