الفهرس | Only 14 pages are availabe for public view |
Abstract Over the years, data mining has attracted most of the attention from the research community. Association rules are a Data mining technique that tries to identify intrinsic patterns in large data sets. It has been widely used in different applications, a lot of algorithms introduced to discover these rules. However most of the algorithms used intend to discretize all numeric attributes, which leads to loss of knowledge. Three algorithms are proposed for mining quantitative association rules using artificial immune system and swarm intelligence. The algorithms intend to discover optimized intervals in numeric attributes without the need for the discretization process. Additionally, the algorithms do not need the minimum support and confidence, instead they look for the best support that conform a frequent itemset. The algorithms are tested with both synthetic and real datasets. The results show that the algorithms provide better accuracy when compared to other algorithms used for quantitative rules. |