| 一种基于关联规则的多类标分类算法 |
| |
| 引用本文： | 李宏,李博,吴敏,陈松乔. 一种基于关联规则的多类标分类算法[J]. 控制与决策, 2009, 24(4) |
| |
| 作者姓名： | 李宏 李博 吴敏 陈松乔 |
| |
| 作者单位： | 中南大学,信息科学与工程学院,长沙,410083 |
| |
| 基金项目： | 国家杰出青年科学基金，中南大学博士后科学基金 |
| |
| 摘 要： | 提出了一种基于关联规则的多类标算法(MLAC).利用多类标FP-tree来分解组合生成多类标规则.并通过组合多重关联规则分类器进行分类预测,降低了由高维属性带来的高计算复杂度,有效地提高了算法的性能和效率.针对多类标数据集的实验结果表明,MLAC算法在性能和效率等方面均优干ML-KNN等多类标分类算法.
|
| 关 键 词： | 分类 关联规则 多类标 单类标 组合 |
Multi-label classification algorithm based on association rules |
| |
| LI Hong,LI Bo,WU Min,CHEN Song-qiao. Multi-label classification algorithm based on association rules[J]. Control and Decision, 2009, 24(4) |
| |
| Authors: | LI Hong LI Bo WU Min CHEN Song-qiao |
| |
| Affiliation: | School of Information Science and Engineering;Central South University;Changsha 410083;China. |
| |
| Abstract: | A multi-label classification algorithm based on association rules (MLAC) is proposed, which uses multi-label FP_tree to resolve multi-label data and merger labels with the same attributes to generate multi-label rules. The classification prediction is conducted by assembling the classifiers. So the computational complexity brought by high dimensional attribute decreases while the performance and efficiency increases. The results of subsequent experiment based on multi-label data show that MLAC achieves bett... |
| |
| Keywords: | Classification Association rules Multi-label Single-label Ensemble |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
