偏最小二乘回归
柠檬酸
化学
苹果酸
相关系数
决定系数
近红外光谱
标准误差
分析化学(期刊)
光谱学
食品科学
色谱法
数学
生物
统计
量子力学
物理
神经科学
作者
Jie Yu Chen,Han Zhang,R. Matsunaga
摘要
The potential of near-infrared (NIR) spectroscopy to rapidly determine citric and malic acid contents of raw Japanese apricot (Japanese "ume", also known as the Japanese plum) fruit juice was investigated. In total, 314 raw juice samples with different organic acid compositions were collected over a long growth period, and spectra (1100-1850 nm) of these samples were acquired using an NIR spectrophotometer with a 1-mm path length. Calibrations were performed using a partial least-squares regression method based on a calibration sample set (211 samples), while validations were performed based on a validation sample set (103 samples). The results revealed good agreement between NIR spectroscopy and capillary electrophoresis, including the correlation coefficient (r2), standard error of prediction (SEP), and bias; no statistically (p = 0.05) significant differences were found for these parameters. Moreover, standard deviation ratios of reference data in the validation sample set to the SEP were higher than 3, indicating that NIR spectroscopy may represent an acceptable method for quantitative evaluation of citric and malic acids in raw Japanese apricot fruit juice.
科研通智能强力驱动
Strongly Powered by AbleSci AI