Reducing Cognitive Overload by Meta-Learning Assisted Algorithm Selection
Reducing Cognitive Overload by Meta-Learning Assisted Algorithm Selection
复制标题
通过元学习辅助算法选择减少认知过载
DOI：
10.4018/jcini.2008070107
复制
发表时间：
2006
期刊：
影响因子：
--
通讯作者：
Minxiao Lei
中科院分区：
文献类型：
--
作者：
L. Fan;Minxiao Lei
随着现有数据挖掘算法的爆炸性增长，帮助用户选择最合适的算法或算法组合来解决问题，减少因算法过载而导致的认知过载变得越来越重要。在本文中，我们探索了一种元学习方法，支持用户在数据挖掘建模过程中自动选择最合适的算法。文中详细讨论了元学习方法，并给出了一些初步的实验结果，表明将元学习方法与粗糙集特征约简相结合可以提高混合模型的性能。可以找到数据集的冗余属性。因此，利用属性约简可以加快排序过程，提高排序精度。随着搜索空间的缩小，用户的认知负荷也随之降低
With the explosion of available data mining algorithms, a method for helping user selecting the most appropriate algorithm or combination of algorithms to solve a problem and reducing cognitive overload due to the overloaded algorithms is becoming increasingly important. In this paper, we have explored a meta-learning approach to support user to automatically select most suited algorithms during data mining model building process. The paper discusses the meta-learning method in details and presents some preliminary empirical results that show the improvement we can achieve with the hybrid model by combining meta-learning method and rough set feature reduction. The redundant properties of the dataset can be found. Thus, we can speed up the ranking process and increase the accuracy by using the reduct of properties. With the reduced searching space, users cognitive load is reduced