Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Wine Quality Analysis
Transcript of Wine Quality Analysis
Data Testing Methods
Computing the Coefficient of correlation to test how strong the mathematical relationship between each variable and the quality score is to see how strongly the variable and quality are related.
The wine making process
All wine making processes start with grapes.
Our Data set;
Hypothesis # 1
A higher alcohol content and lower fixed acidity (relative to other wines) tends to equal a higher quality wine. Why is this?
Hypothesis # 2
Higher quality white wine tends to have a lower residual sugar and lower citric acid. Why is this?
1 - fixed acidity
Acids give the sourness or tartness, “a fundamental feature in wine taste.”
Reduced acidity makes for a “flat” taste.
Examples: tartaric, malic, citric, and succinic acids. They are all found in grapes except for succinic.
Agricultural conditions by year affects grape/Wine quality
11 Chemical variables that may affect wine quality
2 - Volatile Acidity
3 - Citric Acid
4 - Residual Sugar
5 - Chlorides
6 - Free Sulfur Dioxide
7 - Total Sulfur Dioxide
8 - Density
9 - pH
10 - Sulfates
11 - Alcohol
Hypothesis # 3
Hypothesis # 4
Hypothesis # 5
Does lower sulfur content make wine higher quality?
Too much volatile acid is undesirable.
The main acid of issue is acetic acid.
This acid is to be distilled from the wine, leaving only fixed acids.
Most citric acid in the grapes is consumed during fermentation.
It is sometimes added to wine to acidify it and give it “freshness.”
However, this can lead to microbial growth. Tartaric acid is sometimes used instead.
The sugar left in the wine from the grapes. These are sweet wines.
Sweet Taste (sugars + alcohols) <= => Acid Taste
(acids) + Bitter Taste (phenols)
Sugar can give sweetness in taste.
Glucose (sugar) in grapes is an “indication of ripeness.”
The Brix scale is used to track sugar development and determine when to harvest.
The saltiness of wine. Chlorides are a result of the grapes used.
Not bound to any compounds in the wine
a.k.a. Sulfites. Harmful for some people in trace amounts.
An anti-microbial agent – limits growth of harmful yeast/bacteria
“Good” yeasts are resistant to SO2
Too much SO2 can kill the “good” yeast.
Pungent odor is a sign of too much.
All SO2 in the wine, bound and free.
A comparison of the weight of a certain volume of wine to an equivalent volume of water.
Describes the acidity.
Lower the pH, the higher the acidity.
It relates to the fixed acidity of the wine.
There are “buffer acids” that do not contribute to acidity but help keep the pH level.
“Sulfates are mineral salts containing sulfur.”
They can be a by product of animal or plant decay as well as industrial processes.
Sulphates may be connected with fermenting nutrition, which affects wine aroma.
Sugars are converted to alcohol in the yeast fermentation.
Too much alcohol compared to other components leads: “hot” wine
It is acceptable for fortified wines.
California Zinfandel's have 17% alcohol, which is very high for wine.
Waterhouse Lab :http://waterhouse.ucdavis.edu/
Aroma Dictiory: http://www.aromadictionary.com/articles/salt_article.html
Wisconsin Dept. Health Services: https://www.dhs.wisconsin.gov/chemical/sulfates.htm
This question is important for understanding wine standards. Does the alcohol balance the flavor? Or do we as people just really like alcohol?
This questions is just interesting to know. Personally, I didn't know sulfates were an ingredient in wine. It is important to know how all of these seemingly unorthodox ingredients constitute good wine, because everyone likes good wine.
It is important to know what constitutes a high quality white wine. Wine is social. And the more you know about it, the more you can talk about it at social gatherings.
Are the quantities of ingredients the same for both white and red wine? Why or why not?
The difference between white and red wine is important to know so that you can choose the appropriate color for the appropriate function. Again, wine is social, and knowing more about wine makes you look cultured to your peers.
Generally, red wine has a higher alcohol content then white wine. Why is this, since we have already seen that higher alcohol content tends to mean a higher quality?
Why is this important? It's always important to know how much alcohol is in your drink.
To test the first hypothesis, we will likely be constructing a heat map that compares the alcohol content to the fixed acidity and uses the quality as the color.
fixedAcidityAndAlcohol = ggplot(wineQualityWhite, aes(x=fixed.acidity, y= alcohol)) + geom_point(aes(fill = quality, color=quality, size =2))
fixedAcidityAndAlcohol + scale_color_gradient(low="white", high="red")
To test our second hypothesis, we will use a method largely similar to the first. We will construct a heat map comparing residual sugar and citric acid using quality as the color.
citricAcidAndResidSugar = ggplot(wineQualityWhite,aes(x=citric.acid, y= residual.sugar)) + geom_point(aes(fill= quality, color = quality, size =2 ))
citricAcidAndResidSugar + scale_color_gradient(low="white", high="red")
To test our third hypothesis, we will either use a heat map or a scatter plot to show that there is a correlation between sulfur content and quality.
sulfurVsQuality = ggplot(wineQuality, aes(x=free.sulfur.dioxide, y=total.sulfur.dioxide)) + geom_point(aes(fill=quality, color=quality, size =2))
sulfurVsQuality + scale_color_gradient(low="white", high="red")
For our fourth hypothesis, we will construct multiple plots, one for each variable, that uses points from both the white and red wine data sets. After we intuitively examine the plots, we can determine the difference in ingredients.
hist(wineQuality$free.sulfur.dioxide, breaks =15)
hist(wineQuality$total.sulfur.dioxide, breaks =15)
For our fifth hypothesis, we will likely use a histogram to compare the alcoholic contents between the two wines.
While we may consider other methods for testing in the future, for now we feel these testing methods will generate the plots necessary to verify our hypotheses.
First, we import our two data sets as data frames:
wineQualityRed<-read.csv("winequality-red.csv",header = T, sep=";")
wineQualityWhite<-read.csv("winequality-white.csv", header = T, sep=";")
Grapes are harvested and the wilted grapes are removed
Red wines are then fermented within the skin.
White wines, however, are pressed to separate the juice from the skin
The wine is then stored in a barrel to complete the maturation
"Fermentation is is the conversion of sugar, by the interaction
of yeasts into alcohol with a by-product of
carbon dioxide (CO2) gas."
Winemakers also use heat during the fermentation process to extract more flavors during the process. depending on the wine a very exact amount of heat is desired. Too much or too little heat results in poor flavor.
During fermentation the wine is also moved around or "worked" to ensure even flavor throughout the barrel.
"Wine is a simple product of the fermentation of mashed grapes. Its quality may vary greatly due to climate, soil, age of the vines, genetics of the vines, and most importantly, the skill of the winemaker. Every year the climate changes, bringing variations to the final product that are classified as vintages. The soil and genetics of the vines tend to remain the same; however the vines continue to age, yielding less, but more intense fruit, with every passing year."