论文部分内容阅读
为研究卷烟风格与卷烟化学成分之间的关系,应用主成分回归建立卷烟GC-M S数据与卷烟风格评吸值之间的定量模型。对于每种卷烟风格,均建立了4种候选模型,然后选取最佳者。通过最佳模型对测试集样品的30种风格进行预测,有23种风格的预测值与评吸值之间的平均绝对误差小于专家评吸时的最小计分单位,所以定量模型的预测结果可靠。此研究表明化学计量学方法在处理卷烟这种复杂体系时的可用性和有效性。这些定量模型可以作为专家评吸卷烟风格的辅助工具。在定量关系不明确的情况下,应当建立多个候选模型,然后从中选择最佳者。
In order to study the relationship between the cigarette style and the chemical composition of cigarettes, a principal component regression method was used to establish a quantitative model between cigarette GC-MS data and cigarette style rating. For each cigarette style, four candidate models were built and the best ones were chosen. The best model is used to predict the 30 styles in the test sample. The average absolute error between the predicted values and the suction values of 23 styles is less than the minimum scoring unit when the experts are smoking, so the prediction of the quantitative model is reliable . This study shows the usability and effectiveness of chemometric methods in handling such complex systems as cigarettes. These quantitative models can be used as an aid to experts in smoking-style cigarettes. In cases where the quantitative relationship is not clear, multiple candidate models should be established and the best one selected.