基于高效液相色譜的中藥材模式識(shí)別新方法
The New Pattern Recognition Method of Traditional Chinese Medicine Based on HPLC
-
摘要: 該文提出了一種基于高效液相色譜的中藥材模式識(shí)別新方法。此方法采用一種新的自組織模糊神經(jīng)網(wǎng)絡(luò)識(shí)別中藥材。網(wǎng)絡(luò)由2級神經(jīng)網(wǎng)絡(luò)構(gòu)成:第1級為子類自動(dòng)聚類結(jié)構(gòu)自組織模糊神經(jīng)網(wǎng)絡(luò)；第2級為有監(jiān)督子類分類神經(jīng)網(wǎng)絡(luò)。整個(gè)網(wǎng)絡(luò)既有神經(jīng)網(wǎng)絡(luò)的學(xué)習(xí)能力，又有模糊系統(tǒng)的基于規(guī)則的推理能力，特別是對子類的自動(dòng)聚類能力。因?yàn)橹兴幉臅?huì)由于產(chǎn)地、采摘時(shí)間和處理方法的不同而有較大的區(qū)別，所以子類的自動(dòng)聚類能力對中藥材模式識(shí)別尤其有意義。另外整個(gè)網(wǎng)絡(luò)還能適應(yīng)特征維數(shù)不等的模糊模式識(shí)別問題。經(jīng)大量中藥材樣本和實(shí)際測試表明，網(wǎng)絡(luò)抗平移、形變和適應(yīng)新產(chǎn)地藥材的能力都很強(qiáng)，達(dá)到了預(yù)期效果。
-
關(guān)鍵詞:
- 中藥; 模式識(shí)別; 模糊神經(jīng)網(wǎng)絡(luò); 色譜
Abstract: The new pattern recognition method of traditional Chinese medicine based on High Performance Liquid Chromatogram(HPLC) is presented. While a new self-organizing fuzzy neural network is also proposed in order to recognize the traditional Chinese medicine. This network is composed of two levels neural network: the first level is a self-organizing fuzzy neural network which has the automatic clustering structure for sub-class; the second level is supervised sub-class classification neural network. The whole network has not only the learning ability to neural network, but also the logic ability to fuzzy system based on rules, especially the automatic clustering ability to sub-class. The automatic clustering ability for sub-class has great meaning to the pattern recognition of traditional Chinese medicine because the traditional Chinese medicine will be very different under the condition of different place come from, different time picked up and different method dealt with. In addition, the whole network can also adapt to the fuzzy pattern recognition problem when the feature dimensions are not equal. It is manifested after abundant of traditional Chinese medicine samples have been used to test the networks abilities. The results show that the resistance to parallel removal, deformation and the adaptability to the medicines coming from new places are very strong, thus the goals expected are achieved. -
黃建明,郭濟(jì)賢.模式識(shí)別及其在生藥學(xué)領(lǐng)域中的應(yīng)用.天然產(chǎn)物研究與開發(fā),1997,9(3):90-92.[2]顧慧兒,陶珍,趙淑敏等.鹽酸洛美沙星的HPLC測定及同類藥物鑒別.中國醫(yī)藥工業(yè)雜志,1997,28(4):169-171.[3]梅丹,張翠蓮,李大魁等.11種市售曲克蘆丁制劑的產(chǎn)品質(zhì)量評價(jià).中國藥學(xué)雜志,1998,33(10):605 608.[4]Li Yi-bo.[J].et al.. Study on the quality evaluation of Chinese medicine based on ANN. Proc. of the IFAC 2001 on CEFEES, Tianjin.2001,:-[5]Meng Joo Re, Jun Liao, Jianya Liu. Fuzzy neural networks-based quality prediction system for sintering process[J].IEEE Trans. on Fuzzy Systems.2000, 8(3):314-324[6]Wu A, Tam P K S. A fuzzy neural network based on fuzzy hierarchy error approach[J].IEEE Trans.on Fuzzy Systems.2000, 8(6):808-816[7]Beatrice Lazzerini, Francesco Marcelloni. A fuzzy approach to 2-D shape recognition[J].IEEE Trans. on Fuzzy Systems.2001, 9(1):5-16[8]李世雄.小波變換及其應(yīng)用.北京:高等教育出版社,1997:28-36.[9]Chin-Teng Lin, Chia-Feng Juang, Chung-Ping Li. Temperature control with a neural fuzzy inference network. IEEE Trans. on Syst., Man, Cybern.-Part C, 1999, 29(3): 440-451.[10]Broomhead D S, Lowe D. Multivariable functional interpolation and adaptive networks. Complex Syst., 1988, 2(1): 321-355. -
計(jì)量
- 文章訪問數(shù): 2461
- HTML全文瀏覽量: 139
- PDF下載量: 762
- 被引次數(shù): 0
下載:
