IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIS005200909A United States Patent [191 [11] Patent Number: 5,200,909 J uergens [45] Date of Patent: Apr. 6, 1993

[54] METHOD FOR EVALUATION OF A LIQUID, Analytical Chemists, Inc., Arlington, Virginia (Four PARTICULARLY teenth Edition, 1984), pp. 187-188, 221-230. [76] Inventor: John P. Juergens, 3011 Hi1lrnontDr., “Official Methods of Analysis of the Association of Oxford, Miss. 38655 Official Analytical Chemists”, Association of Official Analytical Chemists, lnc., Arlington, Va. (1980), pp. [21] Appl. No.: 629,644 180-181, 513-516, 942-943. [22] Filed: Dec. 7, 1990 Ough et al., “Methods for Analysis of and ”, John Wiley & Sons, New York, 1988, pp. 1-6, Related US. Application Data 42-45, 58-59, 61-62, 66, 69-71, 83, 105, 108, 121, [63] Continuation-in-part of Ser. No. 316,667, Feb. 28, 1989, 125-139, 202, 204-208, 212-213, 217-221. abandoned. Primary Examiner-Edward R. Cosimano [51] 1m. (:1.5 ...... coiN 33/14 Attorney, Agent, or Firm-Staas & Halsey [52] US. Cl...... 364/497; 364/401; 364/403; 364/551.01; 436/24 [58] Field of Search ...... 364/401, 403, 497, 551.01; [57] ABSTRACT 436/24 A method and apparatus for classifying wine by chemi [56] References Cited cally analyzing a wine sample to determine content values indicative of amounts of constituents in the wine PUBLICATIONS sample, evaluating potential interaction among constitu Don Blackburn, “Describing Wine Structure Clearly”, ents in the wine sample, converting the content value of Practical Winery, Grapegrowing and Re each of the constituents to a scaled value adjusted for view, vol. 6, No. 5, Jan./Feb., 1986. interaction with the other constituents, the scale to “Chemistry of Wine Making”, Symposium sponsored which each of the content values is converted being by the Division of Agriculture and Food Chemistry at equal in size, and displaying the scale values so that the 165th Meeting of the American Chemical Society, palatability factors for the sampled wine can be deter Dallas, Tex., Apr. 12-13, 1973, Advances in Chemistry mined by viewing the display. Series 137, pp. 134-150, 185-211. “Of?cial Methods of Analysis of the Association of Official Analytical Chemists”, Association of Official 19 Claims, 12 Drawing Sheets

r111 CHEMICAL ANALYSIS OF SAMPLE

DETERMINE OBJECTIVE SCALE VALUES

{-171 DISPLAY OBJECTIVE SCALE VALUES US. Patent ; Apr. 6, 1993 Sheet 1 of 12 5,200,909

FIG. 1

SWEETNESS /-——-20 DRY ' SWEET

SOFT TART

MELLOW PUCKER o 1.5 10 US. Patent ‘ Apr. 6, 1993 Sheet 2 of 12 5,200,909

FIG. 2

ALCOHOL CONTENT=<8% 11.5% >=15% I I I l V ' I SCALE VALUE 0 5 10

FIG. 3

PH LEVEL > =35 5.4 - = < 3.0 I = a‘ SCALE VALUE 0 5 , 10

FIG. 4 PHENOLSILITER =<120mg 1560mg >=3000mg I ' = w‘ SCALE VALUE 0 5 10

FIG. 5

REDUCING sue/ms 0% 2.5% >-5% I . l I ' l SCALE VALUE 0 5 10

FIG. 6 500v (DERIVED FROM SCALE VALUES OF , 0 5 10 &TANNINS) I I SCALE VALUE 0 5 1o US. Patent Apr. 6, 1993 Sheet 3 of 12 5,200,909

.mm?.28. 2.2:f.r1 . 52%5:: 6EN

(a: mmwzhwwgm mmhwmh (an? om?( US. Patent Apr. 6, 1993 Sheet 4 of 12 5,200,909 FIG. 7A

CHEMICAL ANALYSIS OF SAMPLE

1 I [-161 DETERMINE OBJECTIVE SCALE VALUES 1 [-171 DISPLAY OBJECTIVE SCALE VALUES

FIG. 10A {431 CHEMICAL ANALYSIS OF WINES I r432 DETERMINE A PALATABILITY FACTOR FOR EACH WINE I [-433 STORE PALATABILITY FACTORS AS PART OF INVENTORY WINE DATA BASE US. Patent Apr. 6, 1993 Sheet 5 of 12 5,200,909

FIG 3 SUBROUTINE ' A I [-200 WELCOME~INTRODUCTION SCREEN

20s SELECT BY COLOR OR F000 ? COLOR F000 220 r210 F000 TYPE '2 SELECT COLOR FISH/RED MEAT/ETC. RED/WHITE/BLUSH r225 COLOR = RED/WHlTE/BLUSH

{-240 SELECT PRlCE RANGE PRICE = = LOW NO PREF = MEDIUM <= MEDIUM = HIGH <= HIGH = VERY HIGH <= VERY HIGH I

COUNTRY YES PREFERENCE ? .0PREFERENCE ? r255 {-265 r270 coumnv= SELECT GRAPE= SELECT NO PREF COUNTRY NO PREF GRAPE US/FRANCE/ETC. FROM CHOICES US. Patent Apr. 6, 1993 Sheet 6 0T 12 5,200,909

FIG‘ 9 SUBFIOUTINE B Q9 r300 GENERAL INFO ON RATING SCALES

s05 YES CUSTOMER KNOWS PREFERENCE scoREs '2 [T310 r_ 31 5 CUSTOMER ENTERs PREFERENCE scoREs c?igawgéfégés PI" P" Pa’ PS FOR wINE com:

6 LIST MOST POPULARF320 COLOR wINEs IN INVENTORY.

325 CUSTOMER NO PREFERS SOMETHING SIMILAR TO A WINE f- 350 LISTED ? PROVIDE MORE YES r330 INFORMATION ON SCALES LIST INFORMATION FOR RATING SCALES ON THIS WINE f- 355 CUSTOMER SELECTS 335 PREFERENCE SCORES No MODIFY RATINGS '2 Pb, Pt, Pa, PS f345 YES [-340 SET PREFERENCE CUSTOMER MODIFI ES 6 SCORES PREFERENCE SCORES Pb, Pt, Pa, Ps Pb, Pt, Pa, PS TO RATINGS FOR THIS WINE é) é US. Patent Apr. 6, 1993 Sheet 7 of 12 5,200,909

FIG. 10 SUBROUTINE 0

1-400 lNlTlALlZE REC ARRAY TO REC(i,1)=999 AND REC(i,2)=BLANK [-405 use mvsmonv WINE DATA BASE 445 ' NO END-OF FILE v 410 \ wcoLo?aoclig?spnsF 9 NO ‘ r440 ' SKIP TO NEXT YES 415 RECORD PRICE =RS NO WPRICE OR NOPREF ?

YES ' 42o COUNTRY = WCOUNTRY OR NOPREF '2 YES 425 GRAPE = WGRAPE OR NOPREF ?

YES (-430

435 SCORE < REC(10,1) ? NO YES (-450 CALL D US. Patent ' Apr. 6, 1993 Sheet 8 of 12 5,200,909

FIG. 11

WINE OOLOR WHITE REO BLUSH BODY 2 2 1 AClD/TARTNESS a 2 1 TANNIN/ASTRINGENCY 1 a 2 SWEETNESS a a a

FIG. 13

suBROuT1NEE _ Q? r600 PRINT RECOMMENDATIONS TO sOREEN AND TAPE

e05 REPEAT WITH YES F610 OTHER PREFERENCES CALL G '2 N O [-615 6) CALL G US. Patent FIG.Apr. 6, 121993 Sheet 9 of 12

SUBROUTINE D (DP r500 n = 9

505 scone <= REC(n,1) NO YES [515 n=n+1 f‘510 : 10 n=n-1

f~525 REC(m,1) = REC(m-1,1) NO 520 REC(m,2) = REC(m-1,2) _@= m m = m ' 1 YES [-530 REC(m,1) = SCORE REC(m,2) = wm

RETURN US. Patent " Apr. 6, 1993 Sheet 10 of 12 5,200,909

FIG. 14 SUBROUTINE G (C? r700 n=71'

710 \k YES 720 ‘ REC(n,1) = 999 I NO r730 FOR WID =REC(n,2) WREC = WREC + 1

r740 n=n+1

750 760 NO n>10 YES U.S. Patent ' Apr. 6, 1993 Sheet 11 of 12

FIG. 15 SUBROUTINE F L r805 USE NONJNVENTORY WINE g) ' DATA BASE CALL G YES 345 NO ff’:END-OF-FILE ? COLOR = 7 NO [-840 WCOLOR on NOPREF . SKIP To NEXT YES 815 RECORD PRICE = OR <= NO WPRICE OR NOPREF ?

YES 820 COUNTRY = NO WCOUNTRY OR NOPREF 0 YES 825 GRAPE = NO WGRAPE OR NOPREF ?

YES [-830 SCORE = IWb-Pbl'Cb + IWI-Ptl’Ct + IWa-Pal’Ca + IWs-Psl'Cs ass scone < Rm NO YES r850 CALL D US. Patent ' Apr. 6, 1993 Sheet 12 of 12 5,200,909

FIG. 16

r900 INPUT DEVICE

F910 COMPUTER

r920 DISPLAY 5,200,909 1 2 ment of the characteristics of a liquid, for example, METHOD FOR EVALUATION OF A LIQUID, wine, in which the characteristics are chemically evalu PARTICULARLY WINE ated, including interactions among the chemical constit uents, and the data resulting from the evaluation is con This application is a continuation-in-part of applica 5 verted to a graphic display, with each of the character tion Ser. No. 07/316,667, ?led Feb. 28, 1989, now aban istics being displayed on separate, equal scales. doned. It is also an object of the present invention to provide BACKGROUND OF THE INVENTION a method which provides the quantitative assessment of wine discussed above. 1. Field of the Invention 10 It is a further object of the present invention to pro This invention relates to a method and apparatus for vide a method which will allow consumers to easily evaluating a liquid, particularly wine, based upon con identify preferential wines. stituents therein, and more particularly, to a method and According to the present invention, there is disclosed apparatus for chemically evaluating various properties a method for classifying wine, comprising the steps of: of a liquid and converting the data derived by the evalu chemically analyzing a wine sample to determine a ation to a standard scale so that a consumer may make plurality of content values indicative of amounts of a an objective evaluation of the liquid based on viewing the display. plurality of constituents in the wine sample; evaluating 2. Description of the Related Art potential interaction among constituents in the wine sample; converting the content value of each of the Consumers rely on labels when making consumer 20 decisions to determine the contents of items being pur plurality of constituents to a scaled value adjusted for chased so that they may evaluate whether the item interaction with other constituents, the scale to which appeals to them. Information on wine labels, for exam each of the content values is converted being equal in ple, if provided at all, usually provides only subjective size; and displaying the scale values so that a plurality of and frequently abstract written descriptions of one or 25 palatability factors for the sampled wine can be deter several characteristics of the wine, for example, aroma, mined by viewing the display. Also disclosed are appa ?avor, sweetness, astringency, or body. Occasionally, ratus for performing the above method. additional information is provided regarding the tech These together with other objects and advantages nique used in making the wine, proper serving tempera which will be subsequently apparent, reside in the de ture for the wine, and guidance on the types of food the 30 tails of construction and operation as more fully herein wine will best accompany. after described and claimed, reference being had to the The subjective descriptions noted above are of lim accompanying drawings forming a part hereof, wherein ited value to the great majority of wine consumers, like numerals refer to like parts throughout. particularly the inexperienced wine consumer. The BRIEF DESCRIPTION OF THE DRAWINGS descriptions are developed by individuals who are 35 highly trained and very experienced in wine evaluation. FIG. 1 is an example of a bar chart for graphically Thus, these subjective interpretations of wine charac displaying various characteristics in accordance with a teristics are generally too abstract and sophisticated for ?rst embodiment of the present invention; the average inexperienced consumer. Additionally, FIG. 2 is a graph showing an alcohol scale in accor taste perceptions vary among individuals, even among 40 dance with the embodiment of FIG. 1; experienced judges. Thus, the subjective information FIG. 3 is a graph of an acid/tartness scale in accor appearing on wine labels can vary according to the taste dance with the embodiment of FIG. 1; perception of the judge making the evaluation. FIG. 4 is a graph of a tannin/astringency scale in Furthermore, interactions may occur among some of accordance with the embodiment of FIG. 1; the chemical constituents in wine such that taste percep 45 FIG. 5 is a graph of a sweetness scale in accordance tions may be altered and therefore may not be directly with the embodiment of FIG. 1; related to the absolute quantities of each constituent FIG. 6 is a graph of a body scale in accordance with present. For these reasons, under current labeling prac the embodiment of FIG. 1; tice, even the most extensive description for a particular FIG. 7 is a block diagram of an apparatus for per wine, including that which displays quantitative data or forming the method of the ?rst embodiment; scales based on quantitative data, does not provide the FIG. 7A is a ?owchart diagram of steps for perform consumer with the kind of objective, understandable ing the method of the ?rst embodiment; information necessary to make a con?dent, informed FIG. 8 is a ?owchart diagram of a subroutine for a wine purchase decision. second embodiment of the present invention; 55 FIG. 9 is a ?owchart diagram of a subroutine in ac SUMMARY OF THE INVENTION cordance with the embodiment of FIG. 8 which enables It is an object of the present invention to provide an a consumer to enter preferred rating scale values for easily understandable, quantitative assessment of the wines; characteristics of a liquid, for example, wine. FIG. 10 is a ?owchart diagram of a subroutine which Another object of the present invention is to provide selects a wine from an inventory wine list based upon an easily understandable, quantitative assessment of the the preference score selected by the FIG. 8 and 9 em characteristics of a liquid, for example, wine, in which bodiments; the characteristics of the liquid are evaluated, including FIG. 10A is a ?owchart diagram for storing palatabil interactions among the chemical constituents, and the ity factors as part of an inventory wine database in data resulting from the evaluation is converted to a 65 accordance with the subroutine of FIG. 10; graphic display. FIG. 11 is a chart showing scale weights used for It is an additional object of the present invention to determining wine rating scores in accordance with the provide an easily understandable, quantitative assess embodiment of FIG. 8; 5,200,909 3 4 FIG. 12 is a ?owchart diagram of a subroutine for l2-l3, 1973, Advances in Chemistry Series 137, the ranking selections in sequential order based upon their following relationships have been determined through SCORE values in accordance with the embodiment of empirical research: FIG. 8; 1) Sweetness is directly related to the amount of re FIG. 13 is a ?owchart diagram of a subroutine for ducing (principally and ) in the outputting recommended wines and for updating a wine wine; history ?le in accordance with the embodiment of FIG. 2) Acidity or tartness is directly related to the pH and 8; to a lesser degree, the titratable acids in the wine; FIG. 14 is a flowchart diagram of a subroutine for 3) Astringency is directly related to the amount of updating a wine data ?le in accordance with the em polyphenol compounds (principally the tannins) in the bodiment of FIG. 8; and wine; and FIG. 15 is a ?owchart diagram of a subroutine for 4) Body is principally related to the relative amounts selection of potential wine recommendations in ac of , and tannins in the wine. cordance with the embodiment of FIG. 8. All of the above constituents (reducing sugars, pH FIG. 16 is a block diagram of an apparatus for per and titratable acids, polyphenol compounds and etha forming the method of the second embodiment. nol) are constituents which can be quanti?ed through objective laboratory measurements. Various methods DESCRIPTION OF THE PREFERRED for analyzing wines to determine the above-identi?ed EMBODIMENT characteristic constituents are described in Chemistry of To understand the disclosed method, an understand 20 Wine Making, supra, and Of?cial Methods of Analysis of ing of the basic characteristics evaluated when choosing the Association of O?'icial Analytical Chemists, published wine is necessary. by the Association of Official Analytical Chemists, Inc., In arriving at the disclosed method of wine evalua Arlington, Vir. (14th Edition, 1984). In addition, wine tion, a three-step research process for development of samples can be sent to laboratories for evaluation, for the wine evaluation and labeling system was conducted. 25 example, Scott Laboratories, Inc. in Petaluma, CA; The three steps are: ETS Laboratories in St. Helena, CA; and VINQUIRY 1) determining the most important characteristic in Winemakers Service and Research Laboratory in consumer’s wine preferences and purchase decisions; Healdsburg, Calif. 2) determining whether those characteristics are It was also discovered through empirical research linked to wine constituents which can be quantitatively that interactions may occur among the above constitu determined, and making the quantitative determina ents which alter the perception of the intensity of the tions; and individual constituents and characteristics. For exam 3) converting the quantitative determinations to ple, high levels of polyphenol compounds (principally readily understandable rating scales which correlate tannins) and high acidity each ca reduce the apparent those characteristics important to consumers with the . Furthermore, the interaction be objective laboratory data. tween sugar and polyphenol compounds (principally The most important characteristics in consumers tannins) is reciprocal, that is, high levels of sugar can wine preference and purchase decisions (step I) were reduce the perceived intensity of polyphenol com determined by a review of the literature and by using pounds (principally tannins) and consequently reduce consumer panel research techniques. The results of the 40 the perceived astringency. High acidity can also dimin consumer panel tests reveal that sweetness, tartness ish the perceived body of wine through its in?uence on (acidity), astringency and body are the most important apparent sweetness and by virtue of its palate cleaning characteristics of wine to consumers. The results of the action. These interactions are continuous across the consumer panel test were found to be consistent with respective scales; however, consumers will have vary data published in several reference texts on the sensory 45 ing sensitivities or thresholds of detection for changes in evaluation of wine and food (see, for example, Wines: the intensity of the constituents and characteristics in Their Sensory Evaluation, by Maynard A. Amerine and volved. The effects of these interactions are most no Edward B. Roessler, W. H. Freeman 8t Co., 1976, and ticeable when one interacting constituent is present in Principles of Sensory Evaluation of Food by Manyard A. low to moderate amounts and one or more of the other Amerine, Rose Marie Pangborn and Edward B. Ro 50 interacting constituents are present in relatively high essler, Academic Press, 1965). The literature indicated amounts. that the above characteristics as fundamentally the most Finally, the disclosed method requires conversion of important in evaluating the quality of wine; the con the laboratory results to readily understandable scales sumer research demonstrated that the same characteris and correlation of the important characteristics with tics played a most important role in consumers prefer objective laboratory measurements of the constituents ences and wine purchases. responsible for those characteristics, including adjust The next step was to determine whether the charac ment for interaction among the constituents (step 3). teristics identi?ed in step 1 were linked to quanti?able This is achieved by creating individual scales, each of wine constituents (step 2). According to the above-men 0-10 units, to express the degrees of the characteristics, tioned references and others, including The Fine Wines 60 e.g., sweetness, acidity, astringency, and body, in the of California, Robert S. Blumberg 8t Hurst Hannum, wine. The 0 and the 10 of the scale represent the lowest Dolphin Books, 1973; Alexis Lichine’s Encyclopedia of and highest degree of each characteristic, respectively. Wines and Spirits (in collaboration with William Fi?eld The low ?gure, 0, can represent the lowest value one and with the assistance of Jonathan Bartlett and Jane would expect in an ordinary . For example, Stockwood), Alfred A. Knopf, Inc., 1972; Chemistry of 65 according to the literature, the low value for tannin/as Wine Making, Symposium sponsored by the Division of tringency has been determined to be 120 milligrams Agricultural and Food Chemistry at the 165th Meeting (mg) per liter for an ordinary table wine. Thus, any of the American Chemical Society, Dallas, Tex., Apr. wine having a value of 120 mg per liter of astringency 5,200,909 5 6 or less, receives a 0 rating for the 0-10 scale. On the where S7~=scale value for tannin/astringency; and Tx other end of the spectrum, it has been determined that a equals laboratory value for tannin, expressed as total tannin/astringency value of 3000 mg per liter or greater phenols. is beyond the level found in an ordinary table wine. An elevated sugar level reduces the apparent astrin Thus, a wine having a tannin/astringency value of 3000 gency of wine; therefore, an adjustment in the astrin‘ mg per liter or more receives a 10 rating. Values falling gency scale is necessary when the total between 120 mg and 3000 mg of tannin/astringency are level exceeds 2.0%. In this case, the following alternate given a value between zero and 10 based upon their formula is used to determine the Tannin/Astringency tannin/astringency level. For example, a level of 1560 scale. mg per liter of tannin/astringency, which is halfway If reducing sugar content is greater than 2.0%, then: between 120 mg and 3000 mg, is given a scale rating of ?ve. Formulas, discussed below, were derived to con s1=(0.625-o.oo313Tx)x(1.4-0.2.s,) (4) vert the raw laboratory data into the scale values. In a ?rst embodiment, the scales are presented graphi Where: cally in the form of a bar chart, with numerical values 81: scale value for tannin/astringency ranging from 0-10. FIG. 1 is an example of one such bar Tx=laboratory value for tannin, expressed as total chart, which can be applied to a or incorpo phenols rated in a label, for example. The sweetness scale 20 Sx=laboratory value for the percent of reducing indicates a sweetness level of 0.2. The acidity scale 30 sugars greater than 2.0%). indicates an acidity level of 7.5. The astringency scale 20 When the reducing sugar content of the wine does 40 indicates a tannin level of 1.5. The body scale 50 not exceed 2.0% the correction term is dropped from indicates a body level of 6.3. the equation and the basic equation for astringency The following equations and scales were derived for determination shown above is used. This threshold level determining an alcohol scale, an acid/tartness scale, a is an approximation which was established through tannin/astringency scale, a sweetness scale, and a body 25 empirical research of the inventor. It is likely that some scale. The non-variables were derived based on the consumers may have greater or less sensitivity to the various high and low ranges de?ning an ordinary table interactive effects and may detect changes in the per wine. Obviously, if the ranges are changed, so will the ceptions of astringency at different threshold levels. non-variables. However, this criterion level accounts for most of the alternation in perception due to this interaction. A. Alcohol Scale For determining the alcohol scale, in accordance D. Sweetness Scale with industry and legal standards, the wines having an The accepted range of values on the sweetness scale alcohol content below 8% or above 15%, by volume, for an ordinary table wine are from 0 percent to 5 per are not considered ordinary table wines by legal de?ni cent total reducing sugars. FIG. 5 illustrates an example tion, and thus are off scale. This is graphically shown in of a sweetness scale utilizing the 0-5 percent range. The FIG. 2. The values for equation: basic equation for determining the scale is as follows:

s,41,=t.43x,4x-11.43 (1) SE28,‘ (5) where S,41¢=scale value for alcohol content and Ax where S,=scale value for sweetness; S, equals labora equals the laboratory value of alcohol content in per‘ tory value for the percent of reducing sugar content. cent by volume. In order to adjust for the negative interaction of high acidity and a high tannin level on perceived sweetness, B. Acid/Tartness Scale 45 the following alternate formula is used to determine the FIG. 3 illustrates a sample scale for acid/tartness. For sweetness scale. ordinary table wines, the industry accepted optimum High acidity is de?ned as a pH less than 3.2 and a high range for pH is a pH level of 3.8 down to 3.0. A pH tannin level is de?ned as total polyphenols greater than value above 3.8 is considered undesirable in a wine, as is 2250 mg/liter. If these levels are exceeded, then: a pH below 3.0. The following equation is used to deter mine the acid/tartness scale: s,,;,=47.s-12_s XpHx (2) S,=sca1e value for sweetness where SpH=the scale value for the acidity or tartness of 55 Sx=laboratory value for the percent of reducing the wine and where pl-Ix equals the laboratory value of sugar content the acidity in terms of pH units. pHx=laboratory value for acidity in terms of pH units (less than 3.2) C. Tannin/Astringency Scale Tx=laboratory value for tannin, expressed as total FIG. 4 illustrates a scale for displaying the tannin/as polyphenols (greater than 2250 mg/liter). tringency level of a wine. The range of astringency The respective correction terms for acid and tannin values for an ordinary table wine is from 120 mg per drop from the equation when the criterion levels of liter to 3000 mg per liter (total phenols). The basic equa pH=3.2 and tannin=2250 mg/liter are not exceeded. tion for determining the position on the scale of the When neither of these constituents reach these critical tested wine is as follows: 65 values, the sweetness scale determination reverts to the basic equation shown above. These threshold levels are approximations which were established through empiri cal research of the inventor. It is likely that some con 5,200,909 7 8 sumers will have greater or less sensitivity to the inter FIG. 1. This display allows a consumer to evaluate a active effects and may detect sweetness intensity wine and decide if it has desirable characteristics as changes at different threshold levels. However, these discerned by their individual taste perception merely by criterion levels account for most of the alteration in reading the display. perception due to these interactions. The following examples illustrate the conversion of E. Body Scale raw laboratory measurement data to scale values in accordance with the first embodiment. FIG. 6 is a graph of a body scale in accordance with the embodiment of FIG. 1. The body of a wine is deter EXAMPLE 1 mined primarily by the amount of sugars, alcohol and A dry wine with medium body has the following total phenols (tannins) present in the wine. Therefore, laboratory values: the body scale is derived from the 0-10 scale values for sugar, alcohol and tannins. However, adjustments be made to the body scale when any of these compo Alcohol - 13.2% Acid - pH 3.19 nents (sugars, alcohol and tannins) differ substantially Sugar - 0.1% Tannin - 305 mg/Liter from certain points on their respective scales. To ac commodate these adjustments, there are four different Alcohol = 7.4 Sweetness = 0.2 (corrected for high formulas A-D for body scale (8,) determination, de acidity) pending upon the scale value of tannin and sugars in the Acid/'1'artness = 7.5 wine. 20 Tannin/Astringency = 1.5 Body = 6.3 (corrected for high acidity) Body Scale A: If Tannin Scale (57‘) < 4 and Sweetness Scale (S5) = < 3, then: 5B = ($T+ 5s + $Alc)/3 + (5.41: - 5/1-43) + l3 — STI (7) EXAMPLE 2 Body Scale B: If Tannin Scale (57‘) < 4 and Sweetness Scale 25 ($5) > 3, then: A dry with fairly light body has the follow 58 = (ST + $5 + SAM/3 + (SAIC — 5/l.43) + (3) ing laboratory values: (S5 — 3)'0.5 + l3 — S11

Body Scale C: lfTannin Scale (57') > = 4 and Sweetness Scale (S5) = < 3, then: 30 Alcohol - 12.0% Acid - pH 3.34 55 = (87‘ + $5 + SAM/3 + (5,41: — 5/l.43) + )5 — ST] (9) Sugar - 0.1% ' Tannin 1000 mg/Liter m Body Scale D: lfTannin Scale (57) > = 4 and Sweetness Scale Alcohol = 5.7 ($5) > 3, then: Sweetness = 0.2 $5 = (ST + 5s + SAM/3 + (5.41: — 5/ L43) + (10) Acidfl‘artness = 5.8 (S5 — 3)'0.5 + [5 — STI 35 Tannin/Astringency = 3.8 Body = 4.5 As mentioned previously, acidity can influence the perceived body of wine through its interaction with sugar and directly due to its mouth or palate cleansing EXAMPLE 3 effect. Since the sweetness scale is a component of the A somewhat sweet, but acidic rose or blush wine has body determination, the impact of high acidity through the following laboratory values: sweetness is accounted for. However, an additional direct correction factor is needed for the mouth cleans ing effect of high acid. Since the body scales are derived Alcohol - 11.0% Acid - pH 2.92 from other component scales, the scale value for acidity Sugar - 3.0% Tannin - 470 mg/Liter is used to calculate this correction factor. 45 Scale values: High acidity was de?ned as a pH less than 3.2 which Alcohol = 4.3 Sweetness = 5.5 (corrected for high is equivalent to a scale value of 7.5. Therefore, if the acidity) acidity scale value exceeds 7.5, the result of each of the Acid/tanness = 10 preceding equations is multiplied by the following cor Tannin/Astringency = 1.7 (corrected for high rection term: sugar content) - Body = 3.8 (corrected for high 2-0.133 SP” (11) acidity) where: 55 EXAMPLE 4 Spy: scale value for acidity/tartness A very dry, heavy red wine has the following labora When acidity levels do not exceed a scale value of 7.5 tory values: this term is not used and body determination reverts to the basic formulas above. As with the other corrections factors, this threshold level also is an approximation Alcohol - 13.5% Acid - pH 3.61 which was established through empirical research of Sugar - 0.2% Tannin - 2345 mg/Liter the inventor. It is likely that some consumers may have Scale Values: greater or less sensitivity to the interactive effects and Alcohol = 7.9 Sweetness = 0.36 (corrected for high may detect changes in the body at a different threshold tannin level) level. However, this criterion level accounts for most of Acidfl‘artnss = 2.4 the alteration in perception due to this interaction. Tannin/Astringency = 8.0 As noted above, the scale values for each wine can be Body = 10 displayed in graphic form as shown, for example, in 5,200,909 9 10 The scale values are presented graphically as shown in food is input. The program automatically matches the FIG. 1. The rating shown in FIG. 1 is for the wine in particular food type with the conventionally recog Example 1. nized color wine that goes with that particular food FIG. 7 is a block diagram of an apparatus for per type (steps 220 and 225) if the user inputs a food type. forming the method of the ?rst embodiment. FIG. 7A is Alternatively, if the user has a preference, a ?owchart diagram of steps for performing the method the user inputs the preference at step 205, and the pro of the first embodiment. Testing device 110 is a testing gram selects that color as a limiter at step 210. The user means for testing the wine sample to determine the can chose a price range, or opt not to limit the wine amount of the measurable constituents in the sample choice to any particular price range, at step 230. If the (step 111 in FIG. 7A). For example, testing device 110 consumer chooses not to limit the wine to any price comprises a sweetness tester 120, an acidity tester 130, range, the program indicates NOPREF at step 235 and an astringency tester 140 and a body tester 150. The moves on to a country preference query at step 245. If wine sample is tested by the testing device 110, and the the consumer does want to limit the price to a certain results of the test are input to computer 160. These range, the program goes to select price range step 240, testers can be any conventional testing devices. For at which point the parameters of the range are input by example, the analysis methods described above can be the consumer. After the price range has been de?ned, used and then the results can be manually input to the the program moves on to the country preference query computer 160 via a keyboard. Alternatively, the entire step 245. system could be automated, with testing devices being If the consumer has a preference of country, the directly connected to the computer 160 so as to provide 20 country is input at step 255. Otherwise, the consumer the test results directly to the computer. Computer 160 indicates no country preference (N OPREF) at step 250. performs the necessary calculations to develop the Finally, at step 260, the user has the option of selecting above-mentioned scale values (step 171 in FIG. 7A), a grape preference from a menu (step 270) or opt not to and converts the scale values- to a data format for graph chose a grape type limiter (step 265). If the consumer ically displaying the scales. The formatted data is out 25 chooses not to limit the wine to a grape type, the pro put to label marker 170 (step 171 in FIG. 7A), which gram indicates NOPREF at step 265 and moves on to prints the labels which include the scales thereon. subroutine 13. Alternatively, if the consumer does want In a second embodiment, the rating system is incorpo to limit the wine to a grape type, the program goes to rated into an interactive computer program designed to select grape from choices step 270, at which point the assist consumers in identifying wines with the charac 30 consumer selects a grape preference from a menu. After teristics they desire at the point of purchase. For exam the grape preference is selected, the program moves on ple, the program can be installed in a computer located to subroutine B. in retail outlets for wines. The program is menu driven Once through subroutine A of FIG. 8, the program to allow the consumer to specify the characteristics moves to subroutine B. FIG. 9 is a ?owchart diagram of desired within a given price range for wines stocked by 35 subroutine B which enables a consumer to enter pre that particular retailer. The program also permits the ferred rating scale values for wines if they are known, consumer to specify a wine with which he or she is or, if the rating preference scales are not known, edu familiar as a reference point. The program presents a list cates the consumer on various wine characteristics and of wines with rating scales that possess the desired char their relation to the scales, and then allows the con acteristics input or which are similar to the wines speci 4-0 sumer to chose rating scale values (preference scores). ?ed. Step 300 provides the consumer with general informa If the consumer specifies limits on any of the charac tion on the rating scales, for example, defmitions of the teristics, the program will keep the wine selected within scale ranges and what they mean. In step 305, the con these limits. If the consumer does not specify limits on sumer is asked to input whether or not preference any of the characteristics, the program defaults to pa 45 scores are known. If the consumer answers yes, the rameters which are, for example, plus or minus 1 scale consumer is requested to enter the preference scores in unit for that particular characteristic. For example, a step 310 and the program moves on to subroutine C. If consumer might specify a wine having a sweetness level the consumer does not know his preference scores, the of between 3-6. Alternatively, if the consumer speci?es program moves to step 315 which displays a explana a sweetness level of 4, the program can default to wines 50 tion of the major characteristics for wine color, and step in the 3-5 sweetness range. Of course, the computer 320 which displays a listing of the most popular color could also default so that it picks the precise value speci wines in the inventory. If the consumer decides that ?ed. Any parameter used by a consumer for selecting none of the wines on the list in step 320 are desirable, wine can be included, including the place of origin of the customer inputs “NO” at step 325 and is then pro the wine, the price, the grape type, etc. 55 vided more information on the rating scales at step 350. FIGS. 8-10 and 12-15 are ?owchart diagrams for an The customer is then asked (355) to select preference example of a computer program for implementing the scores for preferred body (Pb), preferred tannin (Pt), wine evaluation and display system of the present in preferred astringency (Pa) and preferred sweetness (Ps). vention. If, however, at step 325, the consumer indicates that FIG. 8 is a ?ow chart diagram of a subroutine (sub one of the wines listed in step 320 is desirable, the cus routine A) which introduces the system to the user and tomer inputs "YES” and is given rating scale scores for allows the user to identify the variables to be used in that particular wine at step 330. At step 335, the con arriving at the wine list of wines having the identi?ed sumer is given the option of modifying the rating scale characteristics. For example, in FIG. 8, the user is ?rst score for that wine (step 340) or selecting the preference introduced to the system by a welcome-instruction 65 scores to be the same as the rating scale scores for the screen (step 200) which contains informational descrip listed wine (step 345). If the consumer chooses to mod tions of the system. The user then proceeds to step 205, ify the rating scale scores for the listed wine at step 335, where a particular color wine, or a particular type of the program goes to step 340, at which point the cus 5,200,909 11 12 tomer modi?es the rating scale scores. After the rating acidity and sweetness characteristics of the wine based scale scores are modi?ed, the program moves on to upon color of the wine, determined by reference to subroutine C. Alternatively, if the consumer chooses to FIG. 11. FIG. 11 is a chart showing the various scale select the preference scores to be the same as the rating weights used in the above equation for the various char scale scores for the listed wine at step 335, the program acteristics of the wines based on their color. Based upon goes to step 345 and sets the preference scores to the wine literature and studies done by the applicant, it has rating scale scores for the listed wine. After the prefer been determined that the body, acidity, tannin and ence scores are set, the program moves on to subroutine sweetness characteristics are of varying importance C. The ?nal result of subroutine B is that preference depending upon the color of the wine. For example, as scores are chosen in accordance with the consumer’s input. Once these preference scores are chosen, the can be seen in FIG. 11, a has a scale weight program moves to subroutine C. of 2 for body, 3 for acidity and sweetness, and l for FIG. 10 is a ?owchart diagram of subroutine C which tannin. The low number, e.g., 1 for tannin, indicates that selects a wine from an inventory wine list based on the for white wine, tannin is of less importance, or is less of preference scores selected by the FIGS. 8 and 9 subrou a factor in evaluating wine than is acidity or sweetness tines. Step 400 is an initialization step. The ?rst value which have scale values of 3. These values are used in REC (i, l)=999 resets a ranking parameter. REC (i, determining the SCORE value so that each characteris 2)=blank is a parameter for the wine record number. tic is given a weighted value of importance roughly Since at the initialization stage, no wine has been com equivalent to its value of importance in perception. A low SCORE indicates that the wine deviates less pared to the preferred parameters, there is no record 20 and it therefore is set to “blank”. The variable “i” repre from the customer preference than a wine having a sents a rank number (1-10 in this example). . higher SCORE. The subroutine of FIG. 9 ranks and Step 405 instructs the program to use an inventory saves the “best ten” (i=l-l0) wines and discards the wine database. The inventory wine database is a stored rest. After a SCORE is determined for the particular list of wines that are presently in inventory. Each wine 25 inventory wine, it is compared with the tenth ranked has a wine record number so that it may be distin record (REC (10, 1)), if such a record exists (step 435). guished from other wines. . Obviously, the ?rst 10 records checked will all make Each wine in the inventory wine database is checked the best 10 until the llth record is checked. If the new to see if it falls within the selected desired parameters. SCORE is greater than the SCORE for the tenth re For example at step 410 the ?rst inventory wine is cord, the inventory wine achieving that SCORE is checked to see if its color (W COLOR) matches the discarded and the next wine in the list is checked (step selected color (COLOR). If the inventory wine color 440). If, however, the inventory wine SCORE is less WCOLOR matches the selected color, or if the con than the tenth wine SCORE, then subroutine D (step sumer has indicated no preference for color (NO 450), shown in FIG. 12 is enabled. PREF), then the inventory wine is next checked for 35 Subroutine D determines where in the selected wine price at step 415. If the inventory wine is not the same ?le the inventory wine should be placed, and then sub as the desired selected color, the subroutine skips to the routine D bumps all of the wines with higher SCORES next record (step 440) and determines if all of the wines up one rank. This discards the previous tenth rank wine. in the inventory wine ?le have been compared with the ’ Step 500 sets 11 equal to 9. This sets up the subroutine preferred parameters (step 445). If all of the wines have D so that it will check the 9th record with the SCORE not been compared, the subroutine goes back to step 410 currently being checked. This check occurs at step 505. and begins the checking process again for the next wine If the SCORE is determined to be less than or equal to on the list. The same sequence takes place for the price the ninth record, one is subtracted from n (for example parameter (step 415), the country parameter (step 420) 9-l=8) at step 510 and then the SCORE is checked and the grape parameter (step 425). If the inventory 45 against the eighth record. This continues until the wine makes it through steps 410-425 without being SCORE is determined to be greater than the SCORE of discarded, this indicates that the wine falls within the the record being checked. At this point the routine goes input preferences of the customer for a color, price, country and grape. At step 430, the wine is rated by to step 515, where the rank is bumped up by l and a value M is set to be equal to the size of the ranking giving it a score (SCORE) based on the following equa tiOn: array, in this example, 10. At step 520 n is compared with M to determine if they are equal. If they are not, the routine goes to step 525. Step 525 moves the list upward until the new selection can be inserted in its proper ranking and step 530 inserts the selection in the 55 list. Once all of the wines are placed in proper rank where Wx=wine characteristics from the inventory order, the subroutine returns to subroutine C and con wine database and Px equals the customer preference tinues until all the records in the inventory wine data for a particular characteristic. FIG. 10A is a ?owchart base have been checked. diagram for storing wine characteristics as part of the Once the end of the inventory wine list is reached, inventory wine database in accordance with the subrou 60 subroutine E, shown in FIG. 13, prints the recom tine of FIG. 10. As in step 111 of the ?rst embodiment, mended wine list (the best ten) to the computer screen each inventory wine is chemically analyzed at step 431. and/or a printer (step 600), and then asks the user if Then, as in step 161 of the ?rst embodiment, wine char there is a wish to repeat the selection process using acteristics are determined for each inventory wine at di?'erent preference values (step 605). step 432 based on the chemical analysis of step 431. 65 In the examples shown in FIG. 13, once a decision is Finally, at step 433 wine characteristics determined at made to repeat or not repeat the wine selection process, step 432 are stored as part of the inventory wine data the subroutine G, shown in FIG. 14, records a tally of base. FIG. Cx is a scale weight for the body, tannin, the wines that were selected by the consumer (steps 610 5,200,909 13 14 or 615). This is a record keeping function which the features and advantages of the invention which fall store owner can later view to determine, for example, within the true spirit and scope thereof. Further, since which wines are the most popular. This information can numerous modi?cations and changes will readily occur be used for marketing, inventory control and sales strat to those skilled in the art, it is not desired to limit the egy. invention to the exact construction and operation illus In FIG. 14, step 700 sets n= 1. The variable “n” rep trated and described, and accordingly all suitable modi resents the rank number of the wine in the “best ten” ?cations and equivalents may be resorted to, falling list. At step 710, the rank is checked to see if it equals an within the scope of the invention. For example, the initialization number, for example, 999. If the rank is disclosed invention is described as being used to evalu equal to the initialization number, the subroutine G ate wine. However, this invention can be used in evalu defaults back to subroutine E (step 720). If the rank ating any liquid that has elements quanti?able by chemi number is not equal to 999, the subroutine goes to step cal analysis, for example, beer, wine coolers, liquor, 730. At step 730, the wine identi?cation number WID is detergent, fuel, etc. identi?ed and a number WREC, which represents the What is claimed is: number of recommendations or potential recommenda 15 1. A method for classifying wine to aid an user in tions for the identi?ed WID, is incremented by 1. This selecting a wine, comprising the steps of: keeps a running count of the number of times the wine (a) chemically analyzing a wine sample to determine was in the “best ten”. At step 740, the rank number n is a ?rst content value based upon an amount of a ?rst incremented by l, and then at step 750 it is determined constituent in the wine sample; if n is greater than 10. If n is greater than 10, the subrou 20 (b) determining an objective scale value indicative of tine reverts back to subroutine E (step 760). If it is less a characteristic of the wine sample based upon the than or equal to 10, the subroutine skips back to step 710 ?rst content value; and and repeats the process. (c) displaying the objective scale value so that the As shown in FIG. 13, if the consumer decides not to user can make an objective evaluation of the wine repeat the wine selection with different preferences, the 25 sample by viewing the objective scale value. subroutine F is executed. Subroutine F, shown in FIG. 2. A method as recited in claim 1, further including 15, performs essentially the same selection process as in the step of chemically analyzing the wine sample to FIG. 9, however, instead of using an inventory wine determine a second content value based upon an database (step 805) a non-inventory wine database is amount of a second constituent in the wine sample, and used. The purpose of the non-inventory wine database step (b) further includes the substep of: is to inform the store owner of wines that would have determining the objective scale value further based been recommended to the consumer had they been in on the second content value and upon an interac the inventory list. This allows the store owner to make tion between the ?rst constituent and the second 'an educated decision regarding which new wines to constituent. stock in inventory. The results of the non-inventory 35 3. A method for classifying wine to aid an user in wine database subroutine F are also input to the G sub selecting a wine, comprising the steps of: routine, which stores the data for later review by the (a) chemically analyzing a wine sample to determine store owner. The steps 805-850 of FIG. 15 correspond content values, each content value being indicative to FIGS. 405-450 of FIG. 9. of an amount of a selected constituent in the wine With the second embodiment of this invention, a wine sample; consumer can easily make a wine decision based upon (b) determining objective scale values respectively quantitative evaluation of the wines available in the indicative of characteristics of the wine sample store. The consumers can change the parameters in based upon the content values and interactions order to experiment with different wines having differ among selected constituents; and ent characteristics, and do so in a controlled fashion so 45 (c) simultaneously displaying the objective scale val that they can re?ne their perceptions. In ues so that the user can make an objective evalua addition, a store owner can keep track of the wines that tion of the wine sample by viewing the objective are good sellers and continue to stock them, and evalu scale value. ate which wines would have been recommended if they 4. A method as recited in claim 3, wherein step (a) had been in the inventory. includes the substeps of determining a content value for FIG. 16 is a block diagram of an apparatus for per tannin Tx and determining a content value for reducing forming the method of the second embodiment. A con~ sugar 5;, and step (b) includes the substep of determin sumer inputs commands at input device 900, which ing an astringency scale value STln accordance with the comprises an input means. Input device 900 can be any following: terminal device, for example a keyboard or touch-pad. 55 if S,>2.0%, then S r: (0.625 --0.003 13 Tx) Input device 900 is coupled to computer 910. Computer ‘(Lb-0.28;); and 910 can be any kind of computer, for example, a mini if Sx<2.0%, then S7=0.625—0.00313Tx. computer, micro-computer, main-frame, etc. Computer 5. A method as recited in claim 3, wherein step (b) 910 receives the commands input from input device 900 includes the sub-step of: and executes the various subroutines, which are stored determining an alcohol scale value, an acidity scale by the computer 900. Display 920, which is coupled to value, an astringency scale value, a sweetness scale computer 910, displays the various screens associated value and a body scale value. with the subroutines. Display 920 can be any type of 6. A method as recited in claim 5, wherein the alcohol display device, for example, a video display terminal, a scale value is determined in accordance with the equa CRT, etc. 65 tion S,41c= 1.43 XAx- l 1.43- where S41; is the alcohol The many features and advantages of the invention scale value and Ax determined in accordance with the are apparent from the detailed speci?cation and thus it equation SS=2Sx where S, is the sweetness scale value is intended by the appended claims to cove all such and S; is the content value for sweetness, the acidity