| a. |
Blocking can be an unnecessary and expensive step in the record linkage pipeline. |
|
|
| b. |
A random forest is a set of set of rules. |
|
|
| c. |
Blocking rules must always be authored manually. |
|
|
| d. |
Active learning of blocking rules minimizes the training data we must provide. |
|
|
| e. |
Matching rules are as expensive to apply as the blocking rules. |
|
|