Using near infrared spectroscopy to authenticate the geographical indicator (gi) of coffee and evaluate the effectiveness of gi on purchase intention

VIETNAM NATIONAL UNIVERSITY HO CHI MINH CITY HO CHI MINH UNIVERSITY OF TECHNOLOGY - - TO PHAN CHIEU DAN USING NEAR-INFRARED SPECTROSCOPY TO AUTHENTICATE THE GEOGRAPHICAL INDICATOR (GI) OF COFFEE AND EVALUATE THE EFFECTIVENESS OF GI ON PURCHASE INTENTION Major: Food Technology ID: 8540101 MASTER THESIS Ho Chi Minh city, July 2022 I H C QU C GIA TP HCM TR NGă I H C BÁCH KHOA - - TÔ PHANăCHIÊUă AN S D NG QUANG PH C N H NG NGO Iă XÁC TH C NGU N G Că ÁNHăGIÁăM Că A LÝ (GI) CÀ PHÊ VÀ HI U QU C AăGIă MUA HÀNG C AăNG I TIÊU DÙNG Chuyên ngành: Công ngh th c ph m Mã s : 8540101 LU NăV NăTH CăS TP H Năụă CHệăMINH,ăthángă07ăn mă2022 NH CỌNGăTRỊNHă TR NGă C HOÀN THÀNH T I: I H C BÁCH KHOA ậ HQG-HCM Cán b h ng d n khoa h c 1: Ti năs ăNguy n Qu c C Cán b h ng d n khoa h că2:ăPhóăGiáoăs ăTi năs ăLêăNguy nă oanăDuy……… Cán b ch m nh n xét 1: Ti năs ,ăNguy năHoàiăH ng ng Cán b ch m nh n xét 2: PGS Ti năs ăHoàngăKim Anh Lu năv năth căs ăđ c b o v t iăTr ngă i h căBáchăKhoa,ă HQGăTp.ăHCM ngàyă13ăthángă07ăn mă2022 Thành ph n H iăđ ngăđánhăgiáălu năv năth căs ăg m: Ch t ch h iăđ ng:ăPhóăGiáoăS ăTi năS ăTơnăN Minh Nguy t Ph n bi n 1: Ti năS ăNguy năHồiăH ngă………………………… Ph n bi nă2:ăPhóăGiáoăS ăTi năS ăHoàngăKimăAnh y viên: Ti năS ăNguy n Qu căC ng…ă………………………………… Th ăkí:ăPhóăGiáoăS ăTi năS ăTr n Th ThuăTrà……… Xác nh n c a Ch t ch H iăđ ngăđánhăgiáăLVăvàăTr ngành sau lu năv năđãăđ CH T CH H Iă ng Khoa qu n lý chuyên c s a ch a (n u có) NGăăăăăăăăăăăăăăTR NG KHOA K THU T HÓA H C TR I H C QU C GIA TP.HCM NGă I H C BÁCH KHOA C NG HÒA Xà H I CH NGH AăVI T NAM c l p - T - H nh phúc NHI M V LU N V NăTH CăS H tên h c viên: TôăPhanăChiêuă an MSHV: 1970372………… Ngày,ătháng,ăn măsinh:ă03/09/1997 N iăsinh:ăQu ngăNgãi…… Chuyên ngành: Công ngh th c ph m Mã s : 8540101 I.TÊNă TÀI: S d ng quang ph c n h ng ngo iăđ xác th c ngu n g căđ a lý (GI) cà phê vàăđánhăgiáăm căđ hi u qu c a GI đ năỦăđ nh mua hàng c aăng i tiêu dùng NHI M V VÀ N I DUNG: 1/ Nhi m v 1: S d ng d li u ph quang ph c n h ng ngo iăđ xác th c ngu n g căđ a lý (GI) cà phê N i dung: Xây d ng mơ hình phân bi t lo i cà phê b ng d li u ph c n h ng ngo i thu b ng máy quang ph c n h ng ngo i c m tay k t h p v i x lý s li uăđaă bi n SIMCA, PLS-DA 2/ Nhi m v 2:ă ánhăgiáăm căđ nhăh ng c a ch ng nh năGIăđ năỦăđ nh mua hàng c aăng i tiêu dùng N i dung: S d ngăđánhăgiáăt p trung (Central Location Test)ăđ đánh giá v ch t l ng bên bên ngồi bao bì cà phê, k t h p v i phân tích Conjoint y u t đ đánhăgiáăđ c y u t nhăh ngăđ năỦăđ nh mua c aăng i tiêu dùng H Chí Minh II NGÀY GIAO NHI M V : 06/09/2021 III NGÀY HOÀN THÀNH NHI M V : 22/05/2022 IV.CÁN B H NG D N (Ghi rõ h c hàm, h c v , h , tên): 1/ Ti n s Nguy n Qu căC ng 2/ PhóăGiáoăs ,ăTi năs ăLêăNguy nă oanăDuy CÁN B H Tp HCM, ngày 01 tháng 07 n mă2022 NG D N TR CH NHI M B NG KHOA K THU T HÓA H C MỌNă ÀOăT O ACKNOWLEDGMENT First of all, we would like to send our sincere appreciation and deepest thanks to our instructors Dr Nguyen Quoc Cuong and Assoc Prof Dr Le Nguyen Doan Duy, for guiding us through the study of this thesis They have always been by our side, willing to sit down and discuss problems or support new knowledge which we have not had a chance to study before, especially with all their heart and dedication With this thank you, we also want to send to all the teachers in Ho Chi Minh City University of Technology, as well as the teachers of the Department of Food Technology - Faculty of Chemical Engineering for their guidance and help to overcome over many difficulties in the process of training and learning during our time at the university After finishing the thesis,ăIăstrongăbelieveăthat:ăắEverythingăwillăbeăokayăinătheă end.ăIfăit'sănotăokay,ăit'sănotătheăend.”ăẮJohn Lennon Besides, because of the lack of knowledge and time, this topic is yet to be optimal, hopefully everyone will contribute to improve the study and the application of classification tools Ho Chi Minh City, July 01, 2022 To Phan Chieu Dan i ABSTRACT Coffeeăisăanăimportantăcommodity,ăaccountingăforă3%ăofăVietnam’ăGDP,ăandătheă Vietnam’ăexportăturnover of coffee has reached over US $3 billion for many years In recent years, although the world economy has experienced many difficult times leadingătoăaădecreaseăinăpurchasingăpower,ăVietnam’săcoffeeăexportsăhaveămaintainedă a very encouraging growth rate, there is an unmet need to protect coffee brand and qualify to prevent the frauds in coffee supply chain Thus, this work proposed the suitability of NIR spectroscopy coupled with chemometrics methods for the nondestructive authentication of Geography indicator coffees For this, GI coffees beans (n=49) and non-GI coffees beans (n=103) by directly analysing without any sample preparation Then, PLS-DA and SIMCA were used to construct the models PLSDA combine with MC, SNV pre-processing constructing model achieved the best result, model accuracy 87.96%, sensitivity 81.25% and specificity 95.23% By the way, SIMCA combine with MC, 2nd Der, SNV pre-processing constructing model achieved the model accuracy just 85.76% Therefore, the proposed methodologies can be useful for both the consumer and regulatory in registration GI for products because it confirms the elevated standard of Dak Lak specialty coffees (GI), preventing fraudulent labeling, friendly with environmental and quick technique, save time and save money Besides, this study also research about the effective of Geography indicator logo to consumer willing to pay This study conducts on 100 consumers participated in a central location test in Ho Chi Minh city Respondents first rated hedonic liking of a cup of black coffee in a blind condition, then evaluated elements of product concepts differing in two extrinsic attributes: region of origin and label claim, before indicating their liking and purchase intent in an informed condition The packaging claim about region of coffee were the strongest drivers for informed liking of coffee, followed by blind liking and price As the same coffee qualify, talk about extrinsic factor:ăGeographyăindicatorălogoăclaimăisăattractiveăconsumerăliking,ăbutăit’sănotăthe importance factor effective to informed liking and purchase intent Consumer awareness about GI is not high to motivate their willing to pay GI coffee with price ii 260.000 VND/ kg Thus, in order to enhance the effectiveness of Geography Indicator to consumer awareness, we need more action to educate consumer about what is Geography Indicator and how it effective to coffee qualify and prevent the fraud of coffee iii TÓM T T LU NăV N Cà phê m t m t hàng quan tr ng, chi m 3% GDP c a Vi t Nam kim ng ch xu t kh u cà phê c a Vi t Nam đãăđ t t đôălaăM nhi uăn m.ăTrongă nh ngăn măg năđây,ăm c dù kinh t th gi i tr i qua nhi uăgiaiăđo năkhóăkh năd n đ n s c mua gi mănh ng xu t kh u cà phê c a Vi t Nam v năduyătrìăđ t ngătr b o v th ct cđ ngăđángăkhíchăl ,ădoăđóăđ đ m b o chu i cung ng cà phê c n có bi n pháp ngăhi uăcàăphêăvàăng năch n hành vi gian l n v ch tăl ng s n ph m Nghiên c u nh m s d ng ph c n h ng ngo i NIR v iăcácăph ngăphápăx lý s li uăChemometricsăđ xác th c b o v th ngăhi u c a càăphêăđ t ch d năđ a lý (GI) vùng Buôn Mê Thu t Trong nghiên c u, s m uăcàăphêănhânăđ t chu n GI (nă=ă49)ăvàăcàăphêănhânăkhôngăđ t chu n GI (n = 103) b ng cách phân tích tr c ti p mà khơng c n phá h y m u.ăSauăđó,ăPLS-DAăvàăSIMCAăđ c s d ngăđ xây d ng mơ hình Mơ hình k t h p PLS-DA v i ti n x lý MC, SNV cho k t qu t t nh t, đ xác c aămơăhìnhă87,96%,ăđ nh yă81,25%ăvàăđ đ c hi u 95,23%.ă th i, mơ hình k t h p SIMCA v i ti n x lý MC,ă2ndăDer,ăSNVăđ tăđ xác c a mơ hình ch 85,76%.ăDoăđó,ăcácăph ng ng căđ ngăphápăđ xu t có th h u ích cho c iătiêuădùngăvàăc ăquanăqu n lý vi căđ ngăkỦăGIăchoăs n ph m kh ng đ nh tiêu chu n ch tăl nhãnămác,ăph ng c aăcàăphêăđ c s nă ngăphápăthânăthi n v iămôiătr k L k (GI),ăng a gian l n ng, k thu t nhanh chóng, ti t ki m th i gian ti n b c Nghiên c uăc ngăđánhăgiáă nhăh baoăbìăđ năỦăđ nh mua c aăng ng c a hình nh logo GI i tiêu dùng Nghiên c u th c hi n 100ăng đánhăgiáăt p trung t i Thành ph H ChíăMinh.ăNg i, i th s đánhăgiáăm căđ yêu thíchătheoăthangăđi mă5;ăđ uătiên,ălàăđánhăgiáăc m quan v ch tăl ng cà phêăđenăphaă phin,ăsauăđó,ăđánhăgiáăn i dung bao bì, cu iăcùngălàăđánhăgiá c m quan s n ph m khiăđãăđ c thơng tin bao bì K t qu nghiên c u cho th y, thông tin xu t s c a cà phê bao bì nhăh sauăđóălàăch tăl ng nhi u nh tăđ n m căđ yêu thích c a s n ph m cà phê, ng c m quan giá c c a s n ph m V i ch tăl ng cà phê, nói v y u t bên ngồi bao bì: thơng tin ch d năđ aălỦă(GI)ăcóătácăđ ng tích c c đ n m căđ yêu thích c aăng i tiêu dùng, nhiên không ph i y u t quan tr ng tácăđ ngăđ năỦăđ nh mua c aăng i tiêu dung Nh n bi t c aăng iv i tiêu dùng v ch d năđ a lỦălàăch aăcao,ădoăđóăch aăđ đ ng l căđ ng i tiêu dùng s n sàng mua s n ph m cà phê có ch d năđ a lý v iăgiáă260.000ăvnđ/kg.ăDoăđó,ăđ t ngăgiáătr c a cà phê có ch d năđ a lý th tr ng, c năt ngănh n th c c aăng d năđ a lý, đ t đóăcóăth phát tri n ch tăl cà phê th tr ng Vi t Nam v i tiêu dùng v ch ng cà phê tránh gi m o ch tăl ng DECLARATION OF AUTHORSHIP I hereby declare that this thesis was carried out by myself under the guidance and supervision of Dr Nguyen Quoc Cuong and Assoc Prof Le Nguyen Doan Duy; and that the work contained and the results in it are true by author and have not violated research ethics The data and figures presented in this thesis are for analysis, comments, and evaluations from various resources by my own work and have been duly acknowledged in the reference part In addition, other comments, reviews and data used by other authors, and organizations have been acknowledged, and explicitly cited I will take full responsibility for any fraud detected in my thesis Ho Chi Minh City, July 01, 2022 To Phan Chieu Dan vi Table 4.9 Relative importance of sensory and non-sensory characteristics for hedonic liking and purchase intent: aggregated results Blind liking Region evaluation Label claim Informed liking (%) 21.51 18.27 60.22 Blind liking Region evaluation Label claim Informed liking Price evaluation 53 Purchase intent (%) 22.72 55.33 0.35 9.08 12.52 CHAPTER 5: CONCLUSION In this study, NIR spectroscopy were exploited with different chemometric approach to test the potential of authentication Geography Indicator coffees and nonGI coffees The result for pre-processing spectra shows that for our NIR data set, SNV and 2nd + SNV on mean centering data set, work well for reducing noise and enhancing peaks The study also aims for investigating which model perform best in detecting adulterants from an instance NIR handheld device Internal and external validation indicate that classification models built by PLS-DA (combine with SNV+ MC) could get accuracy 87.96% and SIMCA (combine with 2nd Der + SNV +MC) could get accuracy 85.76% Overall, PLS-DA model is the best among all Besides the automation, the main benefits related to this kind of analytical approach are the objectivity, the non- destructive nature and its rapidity, leading to a cost- effective improvement of the quality assurance of such a key worldwide of food product With appropriate industrialization and once the chemometric model has been properly calibrated, the time elapsed from the acquisition of NIR spectra on unknow samples and their subsequent classification would require just a few seconds Therefore, this method could represent a concrete and effective answer to the need, claimed by coffee producers, industrial manufacturers, as well as by the Food Authority, of affordable, rapid and efficient technologies for the evaluation of food quality and authenticity This study also confirmed the strong effect of extrinsic cues on informed liking found by previous research It contributes to the existing body of knowledge by disentangling the effect of several extrinsic attributes Overall, packaging claim about region of coffee were the strongest drivers for informed liking of coffee, followed by blind liking As the same coffee qualify, talk about extrinsic factor: Geographyă indicatoră logoă claimă isă attractiveă consumeră liking,ă bută it’să notă theă importance factor effective to informed liking and purchase intent Consumer awareness about GI is not high to motivate their willing to pay GI coffee with price 260.000 VND/ kg Thus, in order to enhance the effectiveness of Geography Indicator to consumer awareness, we need more action to educate consumer about 54 what is Geography Indicator and how it effective to coffee qualify and prevent the fraud of coffee 55 SCIENTIFIC WORKS TôăPhanăChiêuă an,ăLêăNguy nă oanăDuy,ăNguy n Qu căC ph c n h ng ngo i NIR k t h p v iăcácăph ngăphápăx lý s li u chemometrics đ xác th cănhanhăcàăphêăđ t chu n ch d năđ a lý chí Nơng nghi p PTNT, vol 13, p10-25, 2022 56 ng,ăắă ng d ng vùng Buôn Mê Thu t”,ăT p REFERENCES [1] A.A.ăMitchell,ăP.F.Dacin,ăắTheăassessmentăofăalternativeămeasuresăofăconsumer expertise”,ăJournal of Consumer Research, vol 23, pp 219ậ240, 1996 [2] A.Biancolillo, F Marini, Comprehensive Analytical Chemistry, Oxford, Elsevier B.V, 2018, pp 69 ậ 103 [3] A.ăBiancolillo,ăT.ăNæs,ăắTheăsequentialăandăorthogonalizedăPLSăregression for multiblockă regression:ă theory,ă examples,ă andă extensions”,ă Data Handling Science Technology, vol 31, pp 157ậ177, 2019 [4] A Giraudo, S.Grassi, F Savorani, G Gavoci, E Casiraghi, F Geobaldo, ắDeterminationă ofă theă geographicală origină ofă green coffee beans using NIR spectroscopyă andă mulvariateă dataă analysis”.ă Food control, vol.3, pp 25-29, 2019 [5] A Kohler, M Zimonja and V Segtnan, [ Martens, Standard Normal Variate, Multiplicative Signal Correction and Extended Multiplicative Signal Correction Preprocessing in Biospectroscopy, Comprehensive Chemometrics: Chemical and Biochemical Data Analysis, Oxford, Elsevier, 2019, pp 139 ậ 161 [6] A Rinnan,ăF.ăVanădenăBerg,ăS.B.ăEngelsen,ăắReviewăofătheămostăcommonăpreprocessing techniques for near-infraredăspectra”,ăAnalytical Chemistry, vol.28, pp 1201-1222, 2009 [7] A.ăSavitzky,ăM.J.E.ăGolay,ăắSmoothingăandădifferentiationăofădataăbyăsimplifiedă leastăsquaresăprocedure”,ăAnal Chem, vol.11, no.44, pp 1906- 1907, 1972 [8] A.ăZellner,ăắAnăefficient method of estimating seemingly unrelated regressions andătestsăforăaggregationăbias”.ăJournal of the American Statistical Association, vol 57, p.348, 1962 [9] C.C.ăChangăắTheăinterplayăofăproductăclassăknowledgeăandătrialăexperienceăină attitude formation”,ăJournal of Advertising, vol.33, no.1, pp.83ậ92, 2004 [10] C.W Park, V.P Lessig, ắăFamiliarity and its impact on consumer decision biases and heuristics”,ăJournal of Consumer Research, vol 8, no 2, pp 223, 1981 57 [11] D Ballabio, V Consonni,ă ắClassificationă toolsă ină chemistry.ă Partă 1:ă Lineară models PLS-DA”,ăAnalytical Methods, vol 5, pp 3790-3798, 2013 [12] D B Hibbert, Chemometric Analysis of Sensory Data, Comprehensive Chemometrics: Chemical and Biochemical Data Analysis, Oxford, Elsevier, 2009, Vol 4, pp 397 ậ 420 [13] D.ă Cozzolino,ă ắAnă overviewă ofă theă useă ofă infraredă spectrocospyă andă chemometricsă ină authenticityă andă traceabilityă ofă cereals”,ă Food Research International, vol 60, pp 262-265, 2014 [14] D L Wetzel, Near-infrared reflectance analysis of major components in foods, in G Charalambous and G Inglett (eds), Instrumental Analysis of Foods ậ Recent Progress, Vol London: Academic Press, 1983, pp 183 ậ 202 [15] D M Vivien,ă ắAsean Intellectual Property Rights”,ă A booklet on Asean geographical indications procedure and products, vol.1, pp 15-30, 2020 [16] F.ă Gaetano,ă Santeramo,ă E.ă Lamonac,ă ắEvaluationă ofă geographicală labelă ină consumers’ădecision-making process: A systematic review and meta-analysis”,ăă Food Research International, vol 131, pp 108995, 2020 [17] JonnyăForsyth,ăắAă yearăofăinnovationăinăcoffeeă2021”,ă Mintel Food & Drink, 2021 [18] JonnyăForsyth,ăắGlobal: new coffee launches, by key macro-trends, 2010-2020”, Mintel Food & Drink, 2021 [19] J.Paŕ, J.B́langer, Instrumental methods in food analysis, Amsterdam: Elsevier, 1997, pp 94 ậ 102 [20] J.P.ăPerrouty,ăF.ăd’Hauteville,ăL.ăLockshin,ăắTheăinfluenceăofăwineăattributesăonă region of origin equity: An analysis of the moderating effect of consumer’să perceivedăexpertise”,ăAgribusiness, vol 22, pp 323ậ341, 2006 [21] J.W Alba, & J.W Hutchinson,ăắDimensionsăofăconsumerăexpertise”,ăJournal of Consumer Research, vol.13, no.4, pp 411ậ454, 1987 58 [22] K L A.Taynna, N Oliveria, D.S.F David, et al,ă ắNon- destructive authentication of Gourmet ground roasted coffees using NIR spectroscopy and digitalăimages”,ăFood chemistry, vol 135(2012), pp 1828-1835, 2021 [23] L Carbonell, L.Izquierdo, L Carbonell et al.,ăắSegmentationăofăfoodăconsumers according to their correlations with sensory attributes projected on preference spaces”,ăFood Quality and Preference, vol.19, no.1, pp 71ậ78, 2008 [24] L Massart, B.G M Vandeginste, L M C Buydens, S De jong, P J Lewi, and J Smeyers-verbeke,Xvii (1867), Data Handling in Science and Technology Volumn 20A, Elsevier, Amsterdam, 1867, vol 38 [25] L.ăVaĺre,ăTormodNæs,ăắConjointăAnalysisăinăSensory and Consumer Science: Principles,ă Applications,ăandăFutureă Perspectives”,ă Food Science, Technology and Nutrition, pp.485-529, 2018 [26] M Bylesjo, M Rantalainen, Comprehensive Chemometrics: Chemical and Biochemical Data Analysis, Oxford, Elsevier, 2009, Vol 2, pp 131 ậ 134 [27] M.ăDhanoa,ăS.ăLister,ăR.ăSanderson,ăR.ăBarnes,ăắTheălinkăbetweenămultiplicative scatter correction (MSC) and standard normal variate (SNV) transformation of NIRăspectra”,ăJournal of Near Infrared Spectroscopy, vol.2, no.3, pp.43 ậ 47, 1994 [28] M Forina, M Casale, P Oliveri, Application of Chemometrics to Food Chemistry, Comprehensive Chemometrics: Chemical and Biochemical Data Analysis, Oxford, Elsevier, 2009, vol 2, pp.75 ậ 126 [29] M Manfrin, Atencio, J M Engracia, Giraldi, et al,ăắAăcupăofăblackăcoffeeăwithă GI, please! Evidence of geographical indication influence on a coffee tasting experiment”,ăăPhysiology & behavior, vol 245, pp 113671, 2021 [30] MoLi, Z Qian, B Shi, J Medlicott, A East,ăắEvaluatingătheăperformanceăofăaă consumeră scaleă SCiO™ă moleculară sensoră toă predictă qualityă ofă horticulturală products”,ăPostharvest Biology and Technology, vol 145, pp.183-192, 2018 59 [31] M Siegrist, M.E Cousin, Expectations influence sensory experience in a wine tasting, Appetite, 2009, vol 52(3), pp 762ậ765 [32] M Wedel, W.A Kamakura, Market segmentation: Conceptual and methodological foundations, Boston: Kluwer Academic, 2019 [33] M Zeaiter, D Rutledge, Comprehensive Chemometrics: Chemical and Biochemical Data Analysis, Oxford, Elsevier, 2009, vol 3, pp 123 [34] N.ă ăThànhă,ăC s ph 2011 ng pháp ph ng d ng hóa h c, Hà N i, [35] N.ă Mtimet,ă L.M.ă Albisu,ă ắă Spanishă wineă consumeră behavior:ă Aă choiceă experimentăapproach”,ăAgribusiness, vol 22, no 3, pp 343ậ362, 2006 [36] N.T.MăHang,ăN.T.M.ăThuy,ăắThe statusăandăsolutionsăforădevelopingăVietnam’să coffeeăindustry”,ăIndustry and trade magazine, pp.10-20, 2020 [37] O.ăGaltier,ăO.ăAbbas,ăY.ăLe.ăDreau,ăC.ăRebufa,ăJ.ăKister,ăJ.ăArtaud,ăN.ăDupuy,ăắă Comparison of PLS-DA and SIMCA for classification by origin of crude petroleum oils by MIR and virgin olive oils by NIR for different spectral regions”,ăVibrational Spectroscopy, vol 55, pp 132-140, 2010 [38] O.G Mezza-Marquez, T.V Gallardo, G Osorio-Revilla,ăắApplicationăofămidinfrared spectroscopy with multivariate analysis soft independent modelling of class analogies (SIMCA) for the detection of adulterants in minced beef”,ăMeat Sciences, vol 86, pp 511-519, 2010 [39] P Combris, P.Bazoche, E Giraud-H́raud,ăS.ăIssanchou,ăắFoodăchoices:ăWhată we learnăfromăcombiningăsensoryăandăeconomicăexperiments?”,ăFood Quality and Preference, vol.20, no.8, pp.550ậ557, 2009 [40] R.ăDeliza,ăH.J.H.ăMacFie,ăắTheăgenerationăofăsensoryăexpectationăbyăexternală cues and its effect on sensory perception and hedonic ratings:ăAăreview”,ăJournal of Sensory Studies, 11(2), 103ậ128, 1996 [41] R.ă Deliza,ă H.ă MacFie,ă D.ă Hedderleyă ắUseă ofă computer-generated images and conjointă analysisă toă investigateă sensoryă expectations”,ă Journal of Sensory Studies, vol.18, no.6, pp 465ậ486, 2003 60 [42] R Di Monaco, S Cavella, S Di Marzo, et al,ă ắTheă effectă of expectations generatedăbyă brandănameăonătheăacceptabilityă ofădriedăsemolinaăpasta”,ă Food Quality and Preference, vol.15, no.5, pp 429ậ437, 2004 [43] S.ă Cordelle,ă D.ă Piper,ă P.ă Schlich,ă ắOnă theă consistencyă ofă likingă scores:ă Aă validation study run in FranceăandăGermany”,ăFood Quality and Preference, vol 16, pp 493ậ503, 2005 [44] S Mueller, G Szolnoki,ă ắă Theă relativeă influenceă ofă packaging,ă labelling, branding and sensory attribute on liking and purchase intent: consumers differ in their responsiveness”,ă Food quality and preference, vol 21, pp 774-783, 2010 [45] S.ă Mueller,ă L.ă Francis,ă L.ă Lockshin,ă ắTheă relationshipă betweenă wineă liking,ă subjective and objective wine knowledge: Does it matter who is in your ‘consumer’ăsample?”ăăIn Proceedings of fourth international conference of the academy of wine business research, Siena, pp 17ậ19, 2008 [46] S Mueller, L Lockshin, Y Saltman, et al,ăắMessage on a bottle: The relative influenceă ofă backă labelă informationă onă wineă choice”,ă Food Quality and Preference, vol 21, pp 22ậ32, 2010 [47] S.Ră Jaeger,ă J.M.ă Rose,ă ắStatedă choiceă experimentation,ă contextuală influencesă andăfoodăchoice:ăAăcaseăstudy”,ăFood Quality and Preference, vol 19, no 6, pp 539ậ564, 2008 [48] U.ă R.ă Orth,ă K.ă Malkewitz,ă ắHolistică packageă designă andă consumeră brandă impressions”,ăJournal of Marketing, vol 72(3), pp 64ậ81, 2008 [49] (2020) UBND T nhă kăL k Quy tăđ nh Ban hành Quy ch qu n lý, s d ng ch d n đ a lý cà phê Buôn Ma Thu t đ i v i s n ph m cà phê nhân Robusta [50] Y Ozaki, W McClure, A Christy, Near infrared spectroscopy in food science and technology, Hoboken, New Jersey: Wiley-Interscience, 2007, vol 61 APPENDIX A Building a model to authenticate GI coffee and non-GI coffee • Dataset: Coffee NIR Table A.1 Data table of NIR spectra Primary ClassID Huyen Tinh Spectrum 740 + n (0:330,1) Total 152 fixed fixed fixed Numeric samples variables variables variables Character Character Character ID Decriptions: -ClassID: G; N -Huyen: Buon Me Thuot; Ea Hleo; Dak Mil; Chu Prong; Chu Se; Pleiku; Duc Tr ng; Da Lat City -Tinh: Daklak; DakNong; LamDong; GiaLai • Measurement In this context, four different conditions mat occur based on the model prediction, sinceă theă samplesă cană resultă ắtrueă positive”ă (TP),ă ắtrueă negative”ă (TN),ă ắfalseă positive”ă(FP)ăorăắfalseănegative”ă(FN).ăAăseriesăofăstatisticalăparametersăwereăthenă calculated for all the class separated They are: - Sensitivity (SENS), which expresses the model capability to correctly recognize samples belonging to the considered class; SENS= - Specificity (SPEC), which describes the model capability to correctly reject samples belonging to all the other classes; SPEC= - Efficiency (EFF), calculated as the geometric mean of SPEC and SENS EFF = 62 Figure A.1 Data set of NIR spectra 62 B Evaluation the effectiveness of GI label on purchase intent of consumer • Dataset: Consumer evaluation Table B.1 Data table of Consumer evaluation Ma_so_dap_vien Numeric Gioi_tinh Fixed variables, Character Do_tuoi Fixed variables, Character Trinh_do_hoc_van Fixed variables, Character Nghe_nghiep_hien_tai 10 Fixed variables, Character Thu_nhap_HGD Fixed variables, Character Nguoi_quyet_dinh_chinh Fixed variables, Character BUMO Fixed variables, Character MSP Fixed variables, Character Blind_test Numeric Hedonic_liking Numeric Label_claim Numeric Region_evaluation Numeric Purchase_intent Numeric Price _evaluation Numeric Inform_liking Numeric Decriptions: -Gioi_tinh: Nam; N -Do_tuoi: 25-32 tu i; 33-40 tu i -Trinh_do_hoc_van: T t nghi p THCS; T t nghi p PTTH; ang h c ho c t t nghi p trung h c chuyên nghi p; ang h c ho c t t nghi p đ i h c/ Cao đ ng -Nghe_nghiep_hien_tai: Ngh nghi p chuyên môn; T kinh doanh; Buôn bán; Ch doanh nghi p; Chuyên viên có qua tr ng l p đào t o; Nhân viên v n phịng có 63 chuyên môn; Nhân viên bán th i gian/ nhân viên bán hàng; Lao đ ng lành ngh ; Lao đ ng ph thông;N i tr -Thu_nhap_HGD: D i 5,499,000 VND; 5,500,000 – 7,499,000 VND; 7,500,000 –; 13,499,000 VND; 13,500,000 – 22,499,000 VND; Trên 22,500,000 VND -Nguoi_quyet_dinh_chinh: Tôi ng đ nh nhãn hi u ng i quy t đ nh t mua; Tôi ng i khác mua cho -BUMO: - l n/tu n; - l n/tu n;7 l n/tu n -MSP: 267; 625; 802; 324 • Measurement: -Aggregateă(ắtotalăsum”) Analysis: -Compute worst-case running time T(n) for sequence of n operations T(n) -Amortized cost of one operation: = 64 i quy t Ma_so_d Gioi_tinh Do_tuoi Trinh_do_h Nghe_nghiep_h Thu_nhap_HGD ap_vien oc_van ien_tai 1031 1031 1031 1031 1032 1032 1032 1032 1033 1033 1033 1033 1034 1034 1034 1034 1035 1035 1035 1035 1036 1036 1036 1036 1037 1037 1037 1037 1038 1038 1038 1038 Nam Nam Nam Nam N N N N Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam Nam 33 33 33 33 25 25 25 25 33 33 33 33 33 33 33 33 25 25 25 25 25 25 25 25 33 33 33 33 33 33 33 33 40 tu 40 tu 40 tu 40 tu 32 tu 32 tu 32 tu 32 tu 40 tu 40 tu 40 tu 40 tu 40 tu 40 tu 40 tu 40 tu 32 tu 32 tu 32 tu 32 tu 32 tu 32 tu 32 tu 32 tu 40 tu 40 tu 40 tu 40 tu 40 tu 40 tu 40 tu 40 tu T T T T Đ Đ Đ Đ T T T T T T T T Đ Đ Đ Đ T T T T T T T T T T T T t nghi p T L t nghi p T L t nghi p T L t nghi p T L c ho N c ho N c ho N c ho N t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL c ho N c ho N c ho N c ho N t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL t nghi p PL ng lành ng13,500,000 22,499 ng lành ng13,500,000 22,499 ng lành ng13,500,000 22,499 ng lành ng13,500,000 22,499 13,500,000 22,499 13,500,000 22,499 13,500,000 22,499 13,500,000 22,499 ng lành ng7,500,000 13,499, ng lành ng7,500,000 13,499, ng lành ng7,500,000 13,499, ng lành ng7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 7,500,000 13,499, Trên 22,500,000 VN Trên 22,500,000 VN Trên 22,500,000 VN Trên 22,500,000 VN ng lành ng7,500,000 13,499, ng lành ng7,500,000 13,499, ng lành ng7,500,000 13,499, ng lành ng7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 7,500,000 13,499, ng ph th 13,500,000 22,499 ng ph th 13,500,000 22,499 ng ph th 13,500,000 22,499 ng ph th 13,500,000 22,499 Nguoi_quyet_di nh_chinh Tôi ng T T T T T T T T T T T T T T T T T Tôi T T T T T T T T T T T T T i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy i quy BUMO l n/ tu n l n/ tu n l n/ tu n l n/ tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 3-4 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 5-6 l n/tu n 5-6 l n/tu n MSP 267 324 625 802 324 267 802 625 267 625 324 802 267 625 802 324 267 625 802 324 267 625 802 324 267 802 625 324 625 802 324 267 Blind_test Hedonic_lik Label_claim Region_eva Purchase_in Price Inform_liking ing luation tent _evaluation 4 3 4 4 4 5 4 5 4 5 Figure B.1 Data set of consumer evaluation 65 3 5 4 4 4 5 4 4 3 4 4 2 4 4 5 4 4 4 5 4 4 4 3 5 5 4 3 5 4 4 3 2 3 4 4 5 5 5 4 5 5 4 5 4 2 3 3 3 3 3 4 3 2 5 4 4 4 4 4 4 4 4 3 PH N LÝ L CH TRÍCH NGANG H tên:ăTôăPhanăChiêuă an Ngày,ătháng,ăn măsinh:ă03/09/1997ăăăăăăăN iăsinh:ăQu ng Ngãi a ch liên l c: qu n 11, TPHCM QUÁăTRỊNHă ÀOăT O: • i h c: chun ngành cơng ngh th c ph m- i h c Bách Khoa TPHCM (2015-2019) • Cao h c: chuyên ngành công ngh th c ph m- i h c Bách Khoa TPHCM (2019- 2022) Q TRÌNH CƠNG TÁC: Cơng ty c ph n hàng tiêu dùng Masan (2019-nay) 66 ... T Coffee 267 Coffee 324 Coffee 625 PLEIKU Coffee 802 Table 3.3 Sensory expectation test for coffee Coffee 267 Region None Sensory expectation Label None Coffee 324 Coffee 625 Coffee 802 Buon... about coffee type, reputation of region of coffee growth, we need to pay more attention to the geographical origin of products through claim about Geography Indicator 1.4 Geography Indicator (GI) . .. coffee? ??ă[45] Table 3.4 Informed test for coffee evaluation Coffee 267 Coffee 324 Coffee 625 Coffee 802 Buon Me Buon Me Buon Me Buon Me Thuot Thuot Thuot Thuot Informed Region None Buon Me Buon