A soft set approach for association rules mining
In this paper, we present an alternative approach for mining regular association rules and maximal association rules from transactional datasets using soft set theory. This approach is started by a transformation of a transactional dataset into a Boolean-valued information system. Since the "st...
Disimpan dalam:
| Pengarang-pengarang Utama: | , |
|---|---|
| Format: | Artikel |
| Diterbitkan: |
Elsevier Ltd.
2011
|
| Subjek-subjek: | |
| Capaian Atas Talian: | http://eprints.uthm.edu.my/3034/ |
| Penanda-penanda: |
Tambah Penanda
Tiada Penanda, Jadilah orang pertama menanda rekod ini!
|
| Ringkasan: | In this paper, we present an alternative approach for mining regular association rules and maximal association rules from transactional datasets using soft set theory. This approach is started by a transformation of a transactional dataset into a Boolean-valued information system. Since the "standard" soft set deals with such information system, thus a transactional dataset can be represented as a soft set. Using the concept a parameters co-occurrence in a transaction, we define the notion of regular and maximal association rules between two sets of parameters, also their support, confidence and maximal support, maximal confidences, respectively properly using soft theory. The results show that the soft regular and soft maximal association rules provide identical rules as compared to the regular and maximal association rules. |
|---|