量化交易中无需最小支持度阈值的模糊关联规则挖掘

现实量化交易应用中,传统的模糊数据挖掘算法往往需要针对给定的量化交易设定最小支持度阈值,然而,这些方法中存在的普遍问题是很难找到合适的最小支持度阈值,并且因为推导出的规则通常是常识而没有实际的商业意义。为了解决这个问题,提出了一种无需最小支持度阈值的模糊关联规则(fuzzy coherent rule,FCR)挖掘算法。首先将量化交易转换成模糊集,然后通过收集已经生成的模糊集生成候选模糊关联规则,最后计算出列联表并用其检查这些候选模糊关联规则是否满足四项判断准则。如果满足,则可以确定为模糊关联规则。在Foodmart数据集上的实验验证了所提算法的有效性,相比原始模糊关联规则(fuzzy association rules,FAR)挖掘算法,所提的FCR方法能够推导出更多的规则,并且能够在高置信度时推导出更多有用的规则。

Fuzzy Coherent Rule Mining without Minimum Support Threshold Value in Quantitative Transactions

In real-world applications of quantitative transactions, traditional fuzzy data mining approaches usually need setting minimum support threshold value. However, the common problems of those approaches are that an appropriate minimum support is hard to set, and the derived rules usually has no business meanings due to it is exposed common-sense knowledge. To address this problem, an algorithm for mining fuzzy coherent rules (FCR) without minimum support threshold value is proposed. Firstly, quantitative transactions are transformed into fuzzy sets. Then, those generated fuzzy sets are collected to generate candidate fuzzy coherent rules. Finally, contingency tables are calculated and used for checking those candidate fuzzy coherent rules satisfy the four criteria or not. If yes, it is a fuzzy coherent rule. The effectiveness of proposed algorithm has been verified by experiments on the Foodmart dataset. More both rules and helpful rules can be exposed by proposed FCR comparing with primary fuzzy association rules (FAR) approaches.