量子模糊朴素贝叶斯分类算法

Quantum Fuzzy Naive Bayesian Classification Algorithm

以传统朴素贝叶斯算法为基础,研究并提出一种高效、准确的量子模糊贝叶斯分类算法。首先将“模糊集合理论+朴素贝叶斯理论”交叉融合,定义模糊先验概率、模糊条件概率,将朴素贝叶斯推广至模糊朴素贝叶斯,构建模糊贝叶斯模型;其次,将“模糊贝叶斯模型+量子计算”交叉融合,将模糊数据集量子化(编码到量子态上)并设计量子线路,提出一种量子模糊朴素贝叶斯分类算法;最后,将该算法应用到鸢尾花数据集。仿真实验表明,与传统朴素贝叶斯分类算法相比,该算法具有较高的分类效率和准确率。

In today’s era of big data,it is difficult for traditional naive Bayesian algorithms to efficiently and accurately deal with the complexity and uncertainty of big data.Based on the traditional Naive Bayes algorithm,this paper proposes an efficient and accurate quantum fuzzy Bayesian classification algorithm.First,the“fuzzy set theory+naive Bayes theory”is cross-integrated,the fuzzy prior probability and fuzzy conditional probability are defined,and the naive Bayes is extended to fuzzy naive Bayes to construct a fuzzy Bayes model;Secondly,a quantum fuzzy naive Bayesian classification algorithm is investigated and implemented by quantizing fuzzy data sets(encoding to quantum states)and designing quantum circuits.Finally,the algorithm proposed in this paper is applied to the iris dataset.Simulation experiments show that the proposed classification algorithm has higher classification efficiency and accuracy compared with the traditional Naive Bayesian classification algorithm.