Optimized Mining of Potential Positive and Negative Association Rules
| Autor: | Parfait Bemarisika, André Totohasina |
|---|---|
| Přispěvatelé: | Laboratoire d'Informatique et de Mathématiques (LIM), Université de La Réunion (UR), Université d'Antananarivo, Bellatreche, Ladjel, Chakravarthy, Sharma, Univ, Réunion |
| Jazyk: | angličtina |
| Rok vydání: | 2017 |
| Předmět: |
Theoretical computer science
Association rule learning Computer science 020204 information systems 0202 electrical engineering electronic engineering information engineering Structure (category theory) 020201 artificial intelligence & image processing Context (language use) [INFO]Computer Science [cs] 02 engineering and technology Negative association [INFO] Computer Science [cs] ComputingMilieux_MISCELLANEOUS |
| Zdroj: | International Conference on Big Data Analytics and Knowledge Discovery-DaWaK 2017 Big Data Analytics and Knowledge Discovery – 19th International Conference, DaWaK 2017 Big Data Analytics and Knowledge Discovery – 19th International Conference, DaWaK 2017, Aug 2017, Lyon, France. pp.424-432 Big Data Analytics and Knowledge Discovery ISBN: 9783319642826 DaWaK |
| Popis: | The negative association rules are less explored compared to the positive rules. The existing models are limited to the structure of binary data requiring of the repetitive accesses to the context, and the traditional couple support-confiance which is not effective in the presence of the dense data. For that, we propose a new model of optimization by using a new structure of data, noted MatriceSupport, and a new more selective couple, support-\(M_{GK}\). |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|