基于云模型的隶属概念判定算法的改进

在人工智能领域中,李德毅教授等提出的定性定量不确定性转换模型——云模型,把概念的模糊性和随机性完全集成到一起,构成定性数据和定量数据的相互映射。但是,在利用云模型进行隶属概念的判定的算法中,不论对象是否明确属于某个概念,一律进行不确定性判定,过分地强调了模糊性和随机性。这不符合人类的认识规律,而且增加了算法的时间复杂度,甚至导致隶属概念判定的明显错误。该文根据人类认识规律对上述算法进行了改进。对那些可以明确确定属于某概念的对象,不再进行不确定性判定,直接被认定为属于该概念,只对处于多个概念边缘的不能确定属于某个概念的对象进行不确定性判定,使其符合人类的认识规律,降低了算法的时间复杂度。

An Improved Algorithm for Determining Membership Conception Based on Cloud Mode

In the domain of artificial intelligence,uncertainty transformation mode between the qualitative data and the quantitative data put forward by professor Li Deyi etc.,that is,cloud mode,bring the fuzziness and the randomicity of conception together,realizing the reciprocal mapping between the qualitative data and the quantitative data.But,in the algorithm to determine membership conception by making use of cloud mode,all objects would be determined indefinitely,despite that they belongs to certain concept explicitly.Therefore the fuzziness and the randomicity are emphasized excessively,which dose not accord with cognitive rule of the human being and increases time complexity of the algorithm,and even causes the obvious wrong determination of membership conception.This paper improves the above algorithm according to cognitive rule of the human being.Some objects that belong to certain conception explicitly are no longer determined indefinitely but are put to the conception directly.As to the objects which lie in verge of two or multi-conceptions and don't belong to certain conception certainly,they will be determined indefinitely according to the cloud mode. Such an improvement will make the algorithm more identical to cognitive rule of the human being and decrease time complexity of the algorithm.