Principal Component Analysis of Cooked Rice Texture Qualities

March 2008Jourmal of Northeast Agricultural UniversityVol,15 No.1 70-74Principal Component Analysis of Cooked Rice Texture QualitiesLIU Chenghai, ZHENG Xianzhe , and DING NingyeEngineering College, Northeast Agricultural University, Harbin 150030, ChinaAbstract: Texture qualities of cooked rice are comprised of many indexes with the complex relationship. so it is difcult to analyzeand evaluate cooked rice. In this paper, the related indexes of texture properties were conversed into the independent indexes ofprincipal component based on the principal component analysis method. The results showed that the rice kernel types infuenced themeanings of principal components indexes. For long and short rice, the first principal component was comprehensive index. But thesecond principal component was springiness for the short rice, while it was adhesiveness for long rice. Therefore, the first principalcomponent can be used to express the quality of cooked rice with a few of indexes, and the rice type can be recognized according tothe second principal component.Key words: cooked rice, texture quality, principal component analysisCLC number; TS210.1Document code: A Article ID: 1006-8104(2008)-01-0070-05IntroductionMaterials and MethodsRice is one of the most important grain corps of theMaterialsworld, whose taste quality is paid attention by manyLongdao 5 with short type and Songjing 6 with longresearchers and consumers. Taste quality is a sensetype were selected as the experimental samples. Sam-when cooked rice to be chewed, viz stickiness, spring-ples were prepared according to a series of combina-iness, hardness, flavors". The expert fuzzy evaluationtion of chalkiness and crack rates. Chalkiness ratesmethod (EFEM) is the current method to evaluatewere 0%, 5%, 10%, 15%, and crack rates were initial valuethe rice taste quality. The method is high requirement(Longdao 5 with 2.33% and Songjing 6 with 2.00%),for sensory panelists and assessment technique, so25%, 50%, 75%, respectively. In the paper, the full fac-sensory tasting not only wastes time and effort, buttor design of experiments was adopted, and the experi-also is difficult to rapid and effective evaluation formental times were up to 16 groups as shown in Table 1.mass samples'. The texture properties of cooked ricedetermine the rice taste. Therefore, to assess the tasteExperimental equipments and methodsvalue of rice based on the texture qualities of cook-In the paper, rice texture qualities were measured byed rice measured with texture analyzer is an avail-texture analyzer of TA.XT. plus type (England SMSable method with high sensitivity and objectiviy-s.company). The measurement used P/45 probe in TPAThe objective of paper was to determine the efect ofmodel, which was 1 mms' Pre-Test Speed, 0.5 mm.s'the texture qualities of cooked rice on the taste valueTest Speed, 0.5 mm.s' Post-Test Speed, 90% Strain,of rice.and 5.0g Tigger Force The method was that every sam-中国煤化工Received 12 November 2007Supported by Educaion Departmnent of Heiongiang Province in China for the OversMYHCNMHG31002)LIU Chenghai (1982-), malc, Ph. D candidate. engaged in the research of agricultural product processing.* To whom correspondence should be addressed. E-mail: zhengxz@neau edu.cnE-mail: xuebaoenglish@neau.edu.cnLIU Chenghai er al. Principal Component Analysis of Cooked Rice Texture Qualitics.71.ple was measured 5 times, and 3 cooked rice kermelsThe texture results are shown in Table 2.at each time, which were selected randomly and putThe data processing system (DPS) software wasaccording to the same direction.used to analyze experimental data.Table 1 Serial number of dferent chalkiness and crack combination (%)No.Chalkiness rateCrack rateInitial value02105Initial vahue;0122:75141516Table 2 Texture property value of cooked riceVarietyHardncess (g) Adhesivencss (g.see) Springiness Cohesiveness GumminessChewinessResilience3853-2190.3040.3601408.440.1504020-2030.260.3361 3503600.1463 652-2010.2751 2303370.13735220.2570.3241 1432960.1354 067-1040.2690.34714143820.1563919-1790.2540.3311 304323 990-2330.2650.3351 3393540.1394 061-3630.2580.3451 404360.151Longdao 55281-4580.2970.371I 9705930.1794679-3960.3821 793540.1695389-2002811 9400.1784 646-313 .0.2960.37617415150.1666519-3140.3060.3932 5607840.2114 4870.3550.3641 633570.1496 0644870.3440.38623450.1975461.3240.2830.3742 0450.1894363-6490.3130.4171 8180.1614 786-5110.2700.42020107270.1873 9740.2890.38715385364850.2680.40017624623-530.41119176630.1774 296-390.2640.3851660504 205430.2660.39:16775720.1634 0103870.2820.37815244570.153Songing 66 266-3640.3080.409258986215 194- -5670.3072 178810.196114733-4760.2950.40!1 935610.1845 2404820.4182 1977520.202134 807880.322中国煤化工20.1833970.3430.2231s63773770.362MYHCNMHGs.0.2884530.3982 146630.200http: /pblih.eau.edu.cn.72.Journal of Northeast Agricutural UniversityVol. 15 No.1 2008bution law of samples is relatively complex in high-Results and Analysisdimensional space, which increases the complexitiesof analyzing problems.Correlativity analysisTo avoid above-mentioned problerms, principal com-Analyzing correlativity of texture qualities amongponent analysis method was used to analyze the data,every index is shown in Table 2, the results are shownwhose groups of irrelative indexes replaced initial in-in Table 3 and Table 4.dexes68l.According to the results, it is known that there existthe correlativity with different degree among indexes ofPrincipal component analysis (PCA)texture quality for Longdao 5 and Songjing 6, so theThe results of texture qualities of Longdao 5 and Song-information reflected in experiment data is lapped.jing 6 using principal component analysis are shownMoreover, when there are many variables, the distri-in Table5, 6.Table3 Related coefficient of texture indexes of Longdao 5HardnessAdhesivenessSpringiness Cohesiveness GumminessChewinessResilience1.0000-0.59540.41170.73020.99120.92060.9558- 0.5954-0.6945。_0.7411-0.6232-0.7090-0.5921Springiness0.6320).46910.68320.4299Cohesiveness-0.74111.000.76750.82830.8266 .Gumminess0.46910.95470.96100.70900.9176Resilieace0.8266Table 4 Related cofficient of texture indexes of Songjing 6Item0.35690.68670.46260.98330.85320.98720.0644-0.32780.2975-0.08540.30410.21460.7100.68420.65540.51130.72580.52190.7109100000.88940.97220.8599Table 5 Eigenvalue and accumulative percent of correlation matrirPrincipalLongdao 5Songjing 6componentsEigenvalue Percent (%) Accumuative percent (%)EigenvaluePercent (%) Accumulative percent (%)5.4714 .78.16284.698967.12780.934813.354691.51751.419920.284787.41250.30774.396295.91360.63959.135996.5484z0.23363.337499.251098.8652中国煤化工0.02920.416799.667799.64700.02120.303199.9708MHCNMHG99.9069Z0.0020000000.00650.093100.0000E-mail: xuebaoenglish@neau.edu.cnLIU Chenghai et al. Principal Component Analysis of Cooked Rice Texture Qualities.73.Table 6 Eigenvalue and accumulative percent of correlation matrixLongdao5Songing6ItemThe first principalThe second principalcomponentHardness0.39570.35320.44770.1616Adhesiveness-0.33690.44900.08770.7710Springiness0.2897-0.68340.34740.0589Cohesiveness0.3825-0.13690.2792-0.5454Gummniness0.40600.29480.45320.1038Chewiness0.41870.03420.4332-0.2374Resilience0.40080.31580.44750.1065From Table 5, it is known that the first principalpositive eigenvector value. Therefore, the adhesivenesscomponents of Longdao 5 and Songing 6 contain re-of cooked rice is higher,spectively 78.16% and 67.13% information of textureIt is suggested by principal component analysis thatqualities, and the second contain respectively 13.35%texture qualities of cooked rice can be replaced byand 20.28% information. The accumulative percentsthe first two principal components. The first principalof the first two principal components for these twocomponent is comprehensive indexes, including hard-varieties have been over 85%, that is, the first twoness, gumminess, chewiness and resilience, and thereprincipal components may reflect 85% information is also cohesiveness for short type rice. The differencesof 7 initial indexes, therefore, it is feasible to analyzeof information of the second principal componenttexture quality of cooked rice to use the first two prin-from different types have been obtained. The secondcipal components.principal component reflects springiness index forThe results from Table 6 showed that for the Long-short type, and adhesiveness index for long type.dao 5 with short type, the first principal componentcontains 5 factors information, including hardness, Conclusionscohesiveness, gumminess, chewiness and resilience. .The first principal component is higher due to the posi-(1) For the two varieties rice in the experiment, thetive eigenvector value. Therefore, the 5 factors infor-first two principal components contain over 85%mation is also higher. The second principal componentinformation of texture qualities. Therefore, texturewith negative value reflects stringiness level of cookedqualities can be reflected by the first two principalrice. The second principal component is lower, thecomponents.stringiness of cooked rice is lower. However, for Song-(2) Rice kermnel types influence the content of thejing 6 with long type, the first principal componentsecond principal component. It is known accordingcontains 4 factors information including hardness,to above analysis that unknown type rice can begumminess, chewiness and resilience. Positive eigen-confirmed through the result of principal componentvector determine the higher first principal componentanalysis of texture qualities. That is, analyzing thewith the 4 factors information. The result from thesecon中国煤化工lesive is reflected,second principal component reflects adhesiveness.the riif springiness isYHCNMHGThe second principal component is higher due to the reflectcu, 山儿buivigo w ouvss ype.http: /publish.neau.edu.cn.74.Journal of Northeast Agricultural UniversityVol. 15 No.1 20085 Debabandya M, Satish B. Cooking quality and instumental texturalReferencesatributes of cooked rice for dferent milling fractions [0]. Joumal1 Wang z, Gu Y J, Chen G, et al. Rice quality and its afecting factorof Food Engineering, 2006, 73(3);: 253-259.[0]. Molecular Plant Breeding, 2003(2): 231-241.6 Yue TL, Peng B z, Yuan Y H, et al. Modeling of aroma quality2 Zhang H X. Sudies on the mechanical and theological properties ofevaluation of cider based on principal component analysis [].brown rice and cooked rice and discuss of application [D]. Harbin:Transactions of the CSAE, 2007, 23(6): 223-227.Northeast Agricultural University, 2004.7 Nicoletta s, Simona B, Gabriella B, et al. Evaluation of the optimal3 LvJC, Xi x Y. Application of texture analyzer on quality evaluationcooking time of rice by using FT-NIR spectroscopy and anof wheat flour products J. Machinery for Cereals, Oil and Foodelecronic nose [0]. Journal ofCereal Science, 2006, 44(2): 137-143.Processing, 2006(3): 73-74.8 HsuJF, Guo Y L, Yang s y, et al. Congener profiles of PCBs and4 Brenda G L, Elaine T C, Bryan T V, et al. Sensory and instrumentalPCDD/FS in Yucheng victims feen years after exposure to toxicrelationships of texture of cooked rice from slected culivars andrcie-bran oils and their implications for epidemiologice studies切postharvest handling practices [0 Cereal Chem, 2000, 77(1): 64-69.Chemosphere, 2005, 61(19): 1231-1243.中国煤化工MYHCNMHGE-mail: xuebaoenglish@neau.edu.cn