An efficient method for quality control of Fructus Aurantii Immaturus (FAI),a famous traditional Chinese medicine (TCM) was established. A simple and reliable high-performance liquid chromatography-photodiode array de...An efficient method for quality control of Fructus Aurantii Immaturus (FAI),a famous traditional Chinese medicine (TCM) was established. A simple and reliable high-performance liquid chromatography-photodiode array detector (HPLC-DAD) procedure coupled with chemometric methods was developed for fingerprint analysis,qualitative analysis and quantitative determination of this herb. In qualitative and quantitative analyses,heuristic evolving latent projection (HELP) method was employed to resolve the overlapping peaks of the tested samples. Two bioactive components,namely hesperidin and naringin,are confirmed and determined,together with four flavonoids compounds tentatively identified including two new ones. From fingerprint analysis,the fingerprint data were processed with correlation coefficients for quantitative expression of their similarity and dissimilarity. The developed method based on an integration of chromatographic fingerprint and quantitative analysis is scientific,and the obtained results can be applied to the quality control of herb medicine.展开更多
A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decom...A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13% at small training samples and the weaknesses of the conventional methods are overcome.展开更多
基金Project(20875104) supported by the National Natural Science Foundation of ChinaProject(10SDF22) supported by the Special Foundation of China Postdoctoral ScienceProject(201021200011) supported by the Advanced Research Plan of Central South University, China
文摘An efficient method for quality control of Fructus Aurantii Immaturus (FAI),a famous traditional Chinese medicine (TCM) was established. A simple and reliable high-performance liquid chromatography-photodiode array detector (HPLC-DAD) procedure coupled with chemometric methods was developed for fingerprint analysis,qualitative analysis and quantitative determination of this herb. In qualitative and quantitative analyses,heuristic evolving latent projection (HELP) method was employed to resolve the overlapping peaks of the tested samples. Two bioactive components,namely hesperidin and naringin,are confirmed and determined,together with four flavonoids compounds tentatively identified including two new ones. From fingerprint analysis,the fingerprint data were processed with correlation coefficients for quantitative expression of their similarity and dissimilarity. The developed method based on an integration of chromatographic fingerprint and quantitative analysis is scientific,and the obtained results can be applied to the quality control of herb medicine.
文摘A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13% at small training samples and the weaknesses of the conventional methods are overcome.
