| Autor: |
Wei-Ming Huang, Tzung-Pei Hong, Ming-Chao Chiang, Jerry Chun-Wei Lin |
| Jazyk: |
angličtina |
| Rok vydání: |
2019 |
| Předmět: |
|
| Zdroj: |
International Journal of Computational Intelligence Systems, Vol 12, Iss 2 (2019) |
| Druh dokumentu: |
article |
| ISSN: |
1875-6883 |
| DOI: |
10.2991/ijcis.d.190426.001 |
| Popis: |
In the field of fuzzy utility mining, the characteristics of transaction time have been a widely studied topic in data mining. However, using a single-conditional threshold for all items does not suffice to reflect the true properties of items. This paper, therefore, proposes a multi-conditional minimum threshold approach which considers the temporal behavior, the importance in items, and the usage of linguistic terms to mine high temporal fuzzy utility itemsets. The multi-conditional minimum thresholds are utilized to assist users in deciding appropriate standards for itemsets in mining when its contained items have different importance. An extended two-phase model and a corresponding mining approach are designed to handle the temporal fuzzy utility mining problem with the multi-conditional minimum thresholds. Finally, the results from the experimental evaluation show the effectiveness of the proposed approach under different settings. |
| Databáze: |
Directory of Open Access Journals |
| Externí odkaz: |
|
Full text from SpringerLink