摘要
目的建立电子鼻和随机森林算法快速鉴别野生与养殖日本真鲈的分析方法。方法采用来源确定且规格不同的日本真鲈,利用电子鼻中14个金属氧化物传感器获取53份日本真鲈样本(养殖样本25份,野生样本28份)的特征信号,构建得到行×列为53×15(含标签列,野生为1,养殖为-1)的初始特征矩阵。构建随机森林(randomforest,RF)模型,并依据袋外错误率(out-of-bagerrorrate,OOB)对随机森林模型的估计器(决策树)数量和单一决策树最大特征的2个参数进行优化。结果模型最优估计器数为50,最大特征数为14,模型的鉴别准确率达到98.2%。通过该模型,以贡献率为指标,对电子鼻传感器进行了特征筛选和排序,其中S14和S4传感器的贡献率分别为42.9%和36.0%。结论该技术可以快速鉴别野生和养殖日本真鲈。
Objective To establish a rapid identification method for wild and farmed Lateolabrax japonicus by electronic-nose technology and random forest algorithm. Methods Using Lateolabrax japonicus of different sizes with confirmed original materials, feature signals of 53 seabass samples(25 farmed samples, 28 wild samples) were obtained by 14 metal-oxides semiconductor sensors of the electronic-nose. An initial feature matrix formed with row×column as 53×15(labels column included, 1 for wild,-1 for farmed). A random forest model(RF) was constructed, and 2 parameters(estimator number of the RF model and max features of individual decision tree) were optimized according to out-of-bag error rate(OOB). Results The best estimator number was 50, the max feature was 14, and the identification accuracy of the model was 98.2%. According to the model, taking contribution rates as index, the electronic nose senor was selected and ranked, the contribution of S14 and S4 for the identification was 42.9% and 36.0%, respectively. Conclusion This method can rapidly identify the wild and farmed Lateolabrax japonicus.
引文
[1]Claret A,Guerrero L,Gartzia I,et al.Does information affect consumer liking of farmed and wild fish?[J].Aquaculture,2016,(454):157-162.
[2]李智慧,孙永,史建如,等.野生与养殖许氏平鲉品质的比较[J].食品工业科技,2017,38(8):87-91.Li ZH,Sun Y,Shi JR,et al.Comparative of quality of wild-captured and farmed Sebastes schlegeli[J].Sci Technol Food Ind,2017,38(8):87-91.
[3]王玉堂,吕永辉.水产养殖用药与水产品质量安全[J].农业工程,2011,1(3):44-49.Wang YT,Lv YH.Aquaculture drugs and safety of aquatic products[J].Agric Eng,2011,1(3):44-49.
[4]何杰,吴旭干,龙晓文,等.池塘养殖和野生长江水系中华绒蟹扣蟹形态学及生化组成的比较研究[J].水产学报,2015,39(11):1665-1678.He J,Wu XG,Long XW,et al.Comparative studies of morphology and biochemical composition between wild-caught and pond-reared juvenile Chinese mitten crab for Yangtze population[J].J Fish China,2015,39(11):1665-1678.
[5]Wang YV,Wan AHL,Lock E,et al.Know your fish:A novel compound-specific isotope approach for tracing wild and farmed salmon[J].Food Chem,2018,(256):380-389.
[6]Jin J,Deng S,Ying X,et al.Study of herbal tea beverage discrimination method using electronic nose[J].J Food Meas Charact,2015,9(1):52-60.
[7]Tudu B,Ghosh S,Bag AK,et al.Incremental FCM technique for black tea quality evaluation using an electronic nose[J].Fuzzy Inf Eng,2015,7(3):275-289.
[8]Jonsson A,Winquist F,Schnürer J,et al.Electronic nose for microbial quality classification of grains[J].Int J Food Microbiol,1997,35(2):187-193.
[9]Wojnowski W,Majchrzak T,Dymerski T,et al.Electronic noses:Powerful tools in meat quality assessment[J].Meat Sci,2017,(131):119-131.
[10]赵梦醒,丁晓敏,曹荣,等.基于电子鼻技术的鲈鱼新鲜度评价[J].食品科学,2013,34(6):143-147.Zhao MX,Ding XM,Cao R,et al.Identification of Lateolabrax japonicas freshness by electronic nose[J].Food Sci,2013,34(6):143-147.
[11]Qiu S,Wang J.The prediction of food additives in the fruit juice based on electronic nose with chemometrics[J].Food Chem,2017,(230):208-214.
[12]Chen H,Zhang M,Bhandari B,et al.Evaluation of the freshness of fresh-cut green bell pepper(Capsicum annuum var.grossum)using electronic nose[J].LWT-Food Sci Technol,2018,(87):77-84.
[13]王俊,崔绍庆,陈新伟,等.电子鼻传感技术与应用研究进展[J].农业机械学报,2013,44(11):160-167.Wang J,Cui SQ,Chen XW,et al.Advanced technology and new application in electronic nose[J].Trans Chin Soc Agric Mach,2013,44(11):160-167.
[14]Salles T,Gon?alves M,Rodrigues V,et al.Improving random forests by neighborhood projection for effective text classification[J].Inform Syst,2018,(77):1-21.
[15]Vigneau E,Courcoux P,Symoneaux R,et al.Random forests:A machine learning methodology to highlight the volatile organic compounds involved in olfactory perception[J].Food Qual Prefer,2018,(68):135-145.
[16]Huang C,Ma YH,Zhao HB,et al.Spectral classification of asteroids by random forest[J].Chin Astronom Astropys,2017,41(4):549-557.
[17]Edla DR,Mangalorekar K,Dhavalikar G,et al.Classification of EEG data for human mental state analysis using random forest classifier[J].Proced Comput Sci,2018,(132):1523-1532.
[18]中华人民共和国农业农村部渔业渔政管理局.2018年中国渔业统计年鉴[M].北京:中国农业出版社,2018.Bureau of Fisheries,Ministry of agriculture and rurual affairs of the People’s Republic of China.China fisheries statistics yearbook 2018[M].Beijing:China Agriculture Press,2018.
[19]Breiman L.Bagging predictors[J].Mach Learn,1996,24(2):123-140.
[20]Ho T.The random subspace method for constucting decision forests[J].IEEE Trans Patt Anal Mach Intell,1998,20(8):832-844.
[21]Breiman L.Random forests[J].Mach Learn,2001,45(1):5-32.
[22]董师师,黄哲学.随机森林理论浅析[J].集成技术,2013,2(1):1-7.Dong SS,Huang ZX.A brief theoretical overview of random forests[J].JIntegr Technol,2013,2(1):1-7.
[23]Wolpert DH,Macready WG.An efficient method to estimate bagging's generalization error[J].Mach Learn,1999,35(1):41-55.
[24]Genuer R,Poggi J,Tuleau-Malot C.Variable selection using random forests[J].Patt Recogn Lett,2010,31(14):2225-2236.
[25]Guyon I,Elisseeff A.An introduction to variable and feature selection[J].J Mach Learn Res,2003,(3):1157-1182.
[26]Lovell RT.Off-flavors in pond-cultured channel catfish[J].Water Sci Technol,1983,15(6-7):67-73.
[27]Mahmoud MAA,Buettner A.Characterisation of aroma-active and off-odour compounds in German rainbow trout(Oncorhynchus mykiss).Part II:Case of fish meat and skin from earthen-ponds farming[J].Food Chem,2017,(232):841-849.
[28]Chen X,Luo Q,Yuan S,et al.Simultaneous determination of ten taste and odor compounds in drinking water by solid-phase microextraction combined with gas chromatography-mass spectrometry[J].J Environ Sci,2013,25(11):2313-2323.