基因表达式编程(GEP)是一种新颖的遗传算法,是一种高度有效、稳定的随机搜索方法。采用GEP对一系列含氧有机化合物的气相色谱保留指数建立定量结构-保留关系(QSRR)的模型,并与人工神经网络(ANN)预测结果进行比较。GEP和ANN在OV-1固定相上,相关系数R分别为0.9908,0.9892;在SE-54上,相关系数R分别为0.9956,0.9891。结果表明:GEP建立的模型优于ANN,具有良好的拟合度和预测精度。 Gene Expression Programming (GEP) is a novel genetic algorithm, which is a highly effective and stable random search method. The GEP was used to establish a quantitative structure-retention relationship (QSRR) model for gas chromatography retention indices of a series of oxygenated organic compounds and compared with artificial neural network (ANN) predictions. The correlation coefficients R of GEP and ANN on OV-1 stationary phase were 0.9908 and 0.9892 respectively. On SE-54, the correlation coefficients R were 0.9956 and 0.8981, respectively. The results show that the model established by GEP is superior to ANN, and has good fitting degree and prediction accuracy.