One way to improve efficiency of the APRIORI would be to
 |
Prune without checking all k-1 subsets. |
|
|
Join without looping over the entire set, Lk-1. |
This can be done by using hash trees. More details on this can be found in the following lessons.
Other methods to improve efficiency are:
|
|
Speed up searching and matching. |
|
|
Reduce the number of transactions (a kind of instance selection). |
|
|
Reduce the number of passes over data on disk. E.g. Reducing scans via Partition. |
|
|
Reduce number of subsets per transaction that must be considered. |
|
|
Reduce number of candidates (a kind of feature selection). |