| Creator |
|
|
|
|
|
| Language |
|
| Publisher |
|
|
|
| Date |
|
| Source Title |
|
| Vol |
|
| Issue |
|
| First Page |
|
| Last Page |
|
| Publication Type |
|
| Access Rights |
|
| JaLC DOI |
|
| Related DOI |
|
| Related URI |
|
| Relation |
|
| Abstract |
Apriori and its succeeding algorithm DHP are the most known algorithms for mining large itemsets from transaction databases. They are fast when they run once against a transaction database. However, i...f they rerun again and again with different minimum supports, some calculations are done redundantly. With this observation we invented two mining algorithms named FRA and E-FRA, which overcome the above drawback. The algorithms are based on Apriori and DHP. They are different from Apriori and DHP in that new data structures are implemented to improve the performance for one-time run and files in which the useful information found in a run are stored and reused for the afterward run are integrated. This last feature allows better performance for repeated mining with a different minimum support. The paper presents the detailed description of the algorithms and shows some experiments, too.show more
|
Export Mendeley
p081