量子遗传算法优化加权朴素贝叶斯复合语言文本分类

Weighted Naive Bayes Compound Language Text Classification Optimized by Using Quantum Genetic Algorithm

为了提高朴素贝叶斯算法的复合语言文本分类准确度和效率,将加权朴素贝叶斯算法用于复合语言文本分类,采用量子遗传算法对权重参数进行优化;根据贝叶斯定理建立语言文本分类模型,考查样本属性之间的差异对分类结果的影响;然后引入属性权重,形成加权朴素贝叶斯文本分类模型;利用遗传算法对权重参数进行优化,借助量子比特运算提高遗传优化效率,最终得到稳定的复合语言文本分类模型。结果表明,通过合理设置权重个数,量子遗传算法改善了加权朴素贝叶斯算法的文本分类性能,与常用语言文本分类算法对比,该算法具有较高的分类精度和分类效率,在复合语言文本分类中的适用性好。

To further improve the accuracy and efficiency of naive Bayes algorithm for language text classification,weighted naive Bayes algorithm was used for compound language text classification,and quantum genetic algorithm was applied to optimize the weight parameters.A language text classification model was established according to Bayesian theorem,and the effect of differences of sample attributes on classification results were considered.Attribute weights were then introduced to form a weighted naive Bayesian text classification model.The weight parameters were optimized by using genetic algorithm,and the efficiency of genetic optimization was improved by using qubit operation,then a stable compound language text classification model was finally obtained.The results show that quantum genetic algorithm improves the text classification performance of weighted naive Bayes algorithm by reasonably setting the number of weights.Compared with the common language text classification algorithms,this algorithm has high classification accuracy,classification efficiency,and applicability in compound language text classification.