# Explore and summarize data of Red Quality wines
Aim of this project is to find out which chemical property influence the quality of red wines.
# Load all of the packages that you end up using in your analysis in this code
# chunk.
# Notice that the parameter "echo" was set to FALSE for this code chunk. This
# prevents the code from displaying in the knitted HTML output. You should set
# echo=FALSE for all code chunks in your file, unless it makes sense for your
# report to show the code that generated a particular plot.
# The other parameters for "message" and "warning" should also be set to FALSE
# for other code chunks once you have verified that each plot comes out as you
# want it to. This will clean up the flow of your report.
library(ggplot2)
library(tidyr)
library(dplyr)
library(GGally)
library(gridExtra)
library(corrplot)
library(lattice)# Load the Data
wine_df <- read.csv('wineQualityReds.csv', header = TRUE, sep = ',')Univariate Plots Section
Size of dataset
dim(wine_df)## [1] 1599 13The dataset consists of 13 variables and 1599 observations.
Structure of dataset
## 'data.frame': 1599 obs. of 13 variables:
## $ X : int 1 2 3 4 5 6 7 8 9 10 ...
## $ fixed.acidity : num 7.4 7.8 7.8 11.2 7.4 7.4 7.9 7.3 7.8 7.5 ...
## $ volatile.acidity : num 0.7 0.88 0.76 0.28 0.7 0.66 0.6 0.65 0.58 0.5 ...
## $ citric.acid : num 0 0 0.04 0.56 0 0 0.06 0 0.02 0.36 ...
## $ residual.sugar : num 1.9 2.6 2.3 1.9 1.9 1.8 1.6 1.2 2 6.1 ...
## $ chlorides : num 0.076 0.098 0.092 0.075 0.076 0.075 0.069 0.065 0.073 0.071 ...
## $ free.sulfur.dioxide : num 11 25 15 17 11 13 15 15 9 17 ...
## $ total.sulfur.dioxide: num 34 67 54 60 34 40 59 21 18 102 ...
## $ density : num 0.998 0.997 0.997 0.998 0.998 ...
## $ pH : num 3.51 3.2 3.26 3.16 3.51 3.51 3.3 3.39 3.36 3.35 ...
## $ sulphates : num 0.56 0.68 0.65 0.58 0.56 0.56 0.46 0.47 0.57 0.8 ...
## $ alcohol : num 9.4 9.8 9.8 9.8 9.4 9.4 9.4 10 9.5 10.5 ...
## $ quality : int 5 5 5 6 5 5 5 7 7 5 ...Converted the quality variable from type integer to type factor, as it denotes the ratings received by the wine.
Statistical summary of the given dataset
## X fixed.acidity volatile.acidity citric.acid
## Min. : 1.0 Min. : 4.60 Min. :0.1200 Min. :0.000
## 1st Qu.: 400.5 1st Qu.: 7.10 1st Qu.:0.3900 1st Qu.:0.090
## Median : 800.0 Median : 7.90 Median :0.5200 Median :0.260
## Mean : 800.0 Mean : 8.32 Mean :0.5278 Mean :0.271
## 3rd Qu.:1199.5 3rd Qu.: 9.20 3rd Qu.:0.6400 3rd Qu.:0.420
## Max. :1599.0 Max. :15.90 Max. :1.5800 Max. :1.000
## residual.sugar chlorides free.sulfur.dioxide
## Min. : 0.900 Min. :0.01200 Min. : 1.00
## 1st Qu.: 1.900 1st Qu.:0.07000 1st Qu.: 7.00
## Median : 2.200 Median :0.07900 Median :14.00
## Mean : 2.539 Mean :0.08747 Mean :15.87
## 3rd Qu.: 2.600 3rd Qu.:0.09000 3rd Qu.:21.00
## Max. :15.500 Max. :0.61100 Max. :72.00
## total.sulfur.dioxide density pH sulphates
## Min. : 6.00 Min. :0.9901 Min. :2.740 Min. :0.3300
## 1st Qu.: 22.00 1st Qu.:0.9956 1st Qu.:3.210 1st Qu.:0.5500
## Median : 38.00 Median :0.9968 Median :3.310 Median :0.6200
## Mean : 46.47 Mean :0.9967 Mean :3.311 Mean :0.6581
## 3rd Qu.: 62.00 3rd Qu.:0.9978 3rd Qu.:3.400 3rd Qu.:0.7300
## Max. :289.00 Max. :1.0037 Max. :4.010 Max. :2.0000
## alcohol quality
## Min. : 8.40 Min. :3.000
## 1st Qu.: 9.50 1st Qu.:5.000
## Median :10.20 Median :6.000
## Mean :10.42 Mean :5.636
## 3rd Qu.:11.10 3rd Qu.:6.000
## Max. :14.90 Max. :8.000Analyzing variable : Quality
To check out the most frequent quality of wines available in the dataset.
table(wine_df$quality)##
## 3 4 5 6 7 8
## 10 53 681 638 199 18The above stats shows the available number of different qualities of red wine present in the given dataset. The best quality of red wine available in the given dataset is _Quality 8 and it is totally 18 in number, Red wine with Quality 7 is 199 in number.And the most avaliable qualities of red wine are Quality 5 and QUality 6 with 681 and 638 numbers respectively.
p1 <- ggplot(data = wine_df, aes(x=factor(quality))) +
geom_bar()+
labs(x = 'Quality of wines', y = 'Frequency',
title = 'Bargraph of frequency vs Quality of available wines')
p1
So from the above two analysis: it is observed that wines with quality rating 5 and 6 is the most available in the given dataset. Followed by quality 7 and quality 4. Quality 8 followed by quality 3 are the least available.
Analyzing variable : fixed acidity
Description of fixed acidity: most acids involved with wine or fixed or nonvolatile (do not evaporate readily).
ggplot(data = wine_df, aes(x=fixed.acidity))+
geom_histogram(binwidth = 0.5, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(4.5,16), breaks = seq(4.5,16,0.5))+
labs(title = 'Histogram of fixed acidity')## Warning: Removed 2 rows containing missing values (geom_bar).
The above distribution is right-skewed and most of the wines in the given dataset have fixed acidity between 6.5 and 8 g/dm3.
ggplot(data = wine_df, aes(x=fixed.acidity))+
geom_histogram(binwidth = 1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(4.5,16), breaks = seq(4.5,16,1))+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')+
labs(title = 'Faceting the distribution of fixed.acidity based on wine quality')
The above facetted graph, confirms the previous two analysis that the red wines with Quality 5 and 6 are the most available in the dataset and their fixed acidity mostly ranges between 6.5 g/dm3 and 8 g/dm3 respectively. And the best two available qualities of wine Quality 7 and 8 have a very low count and their value of fixed acidity usually varies between 6.5 and 11.5 g/dm3. Also Quality 3 and 4 wines have value of fixed acidity varying between 6,5 and 8.5 g/dm3.
## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 6.700 7.150 7.500 8.360 9.875 11.600
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 4.600 6.800 7.500 7.779 8.400 12.500
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.000 7.100 7.800 8.167 8.900 15.900
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 4.700 7.000 7.900 8.347 9.400 14.300
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 4.900 7.400 8.800 8.872 10.100 15.600
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.000 7.250 8.250 8.567 10.225 12.600The above statistics show the summary of fixed acidity present in different qualities of red wine.
ggplot(data = wine_df, aes(x=quality, y=fixed.acidity))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(6.5,10.5))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title = 'Boxplot of fixed acidity vs Quality of red wine')
The Quality 8 of red wine has median fixed acidity of 8.25 g/dm3 and mean fixed acidity of 8.567 g/dm3. The Quality 7 of red wine has median fixed acidity of 8.8 g/dm3 and mean fixed acidity of 8.872 g/dm3.
The most available Quality 5 of red wine has median fixed-acidity of 7.8 g/dm3 and mean fixed-acidity of 8.167 g/dm3. And Quality 6 of red wine
has median fixed-acidity of 7.9 g/dm3 and mean fixed-acidity of 8.347 g/dm3.
And the available poor qualities of red wine namely Quality 3 and Quality 4 have median fixed acidity of 7.5 g/dm3 each and their mean value of fixed acidity is 8.36 and 7.79 g/dm3 respectively.
The red point shows the mean value of the fixed acidity.
So from the above anlaysis, fixed acidity volume does somewhat improve the quality of the condidered red wine.
Analyzing variable : volatile acidity
Description of volatile acidity: the amount of acetic acid in wine, which at too high of levels can lead to an unpleasant, vinegar taste
ggplot(data = wine_df, aes(x=volatile.acidity))+
geom_histogram(binwidth = 0.1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(0,1.6), breaks = seq(0,1.6,0.1))+
labs(title = 'Histogram of volatile acidity content in the available red wines')
From the above histogram of volatile acidity - most of the wines have volatile acidity in the range of 0.3 to 0.7 g/dm3.
ggplot(data = wine_df, aes(x=volatile.acidity))+
geom_histogram(binwidth = 0.1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(0,1.6), breaks = seq(0,1.6,0.1))+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')+
labs(title = 'Faceting the distribution of volatile.acidity based on wine quality')
Most of the red wines with Quality 8 and Quality 7 have volatile acidity usually ranging from 0.2 to 0.4 g/dm3. Also their distribution is right-skewed,so we can infer from this that good quality red wines have usually low values of volatile acidity.
Red wines of Quality 6 and 5 are roughly normally distributed and their values usually range from 0.2 to 0.7 g/dm3.
Poor quality red wines of Quality 3 and 4 have their values ranging from 0.4 to 1 g/dm3.
## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.4400 0.6475 0.8450 0.8845 1.0100 1.5800
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.230 0.530 0.670 0.694 0.870 1.130
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.180 0.460 0.580 0.577 0.670 1.330
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.1600 0.3800 0.4900 0.4975 0.6000 1.0400
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.1200 0.3000 0.3700 0.4039 0.4850 0.9150
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.2600 0.3350 0.3700 0.4233 0.4725 0.8500The above statistics show the summary of volatile acidity present in different qualities of red wine.
ggplot(data = wine_df, aes(x=quality, y = volatile.acidity))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(0.25,1.05))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title = 'Boxplot of volatile acidity vs Quality of red wine')
The above box plot shows that good quality red wines have lower amounts of volatile acids present in them.
The best available red wine of Quality 8 and Quality 7 have median volatile acidity of 0.37 g/dm3 each and mean volatile acidity of 0.4233 and 0.4033 g/dm3. And as the quality of red wine degrades amount of volatile acidity present in them also increases. The red point shows the mean value of the volatile acidity.
Analyzing variable : Citric Acid
Description of Citric acid: found in small quantities, citric acid can add ‘freshness’ and flavor to wines.
ggplot(data = wine_df, aes(x=citric.acid))+
geom_histogram(binwidth = 0.05, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(0,1), breaks = seq(0,1,0.1))+
labs(title = 'Histogram of Citric acid content in the available red wines')## Warning: Removed 2 rows containing missing values (geom_bar).
The above histogram on amount of citric acid in the red wines is right-skewed.
ggplot(data = wine_df, aes(x=citric.acid))+
geom_histogram(binwidth = 0.05, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(0,1), breaks = seq(0,1,0.1))+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')+
labs(title = 'Faceting the distribution of citric acid based on wine quality')
The above faceted graph shows the amount of citric acid present in different qualities of red wine.
Good red wines of Quality 8 and 7 have citric acid in the range 0.3 to 0.55 g/dm3.
Red wines of Quality 6, 5 and 4 have roughly right skewed distribution and the amount of citric acid in them range from 0 t0 0.6 g/dm3 and their values usually lie around 0.35 to 0.6 g/dm3.
Poor quality red wines of Quality 3 use citric acid in the range from 0 to 0.6 g/dm3.
## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0050 0.0350 0.1710 0.3275 0.6600
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0300 0.0900 0.1742 0.2700 1.0000
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0900 0.2300 0.2437 0.3600 0.7900
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.0900 0.2600 0.2738 0.4300 0.7800
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.3050 0.4000 0.3752 0.4900 0.7600
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0300 0.3025 0.4200 0.3911 0.5300 0.7200The above statistics show the summary of citric acid present in different qualities of red wine.
ggplot(data = wine_df, aes(x=quality, y = citric.acid))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(0,0.55))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title = 'Boxplot of citric acid vs Quality of red wine')
The above box plot shows that good quality red wines have higher amounts of citric acids present in them.
The best available red wine of Quality 8 and Quality 7 have median citric acid of 0.42 g/dm3 and .4 g/dm3 and mean volatile acidity of 0.3911 g/dm3 and 0.3752 g/dm3. And as the quality of red wine degrades amount of citric acid present in them also deceases. The red point shows the mean value of the citric acid.
Analyzing variable : residual sugar
Description of residual sugar: the amount of sugar remaining after fermentation stops, it is rare to find wines with less than 1 gram/liter and wines with greater than 45 grams/liter are considered sweet.
p <- ggplot(wine_df, aes(x=residual.sugar))+
geom_histogram(binwidth = 0.5,color='black',fill='white')+
scale_x_continuous(limits = c(0.5,16), breaks = seq(0.5,16,1))+
labs(title = 'Histogram of residual sugar content in the available red wines')
p
The distribution obtained above is right-skewed distribution. And most of the available red wines have residual sugar in the range of 1.5 to 2.5 grams/liter. Applying log scale to the residual sugar histogram.
ggplot(wine_df, aes(x=residual.sugar))+
geom_histogram(binwidth = 0.1,color='black',fill='white')+
labs(title = 'Histogram of residual sugar content using log_scale')+
scale_x_log10(limits = c(0.8,10), breaks = seq(0.8,10,1))
ggplot(wine_df, aes(x=residual.sugar))+
geom_histogram(binwidth = 0.1,color='black',fill='white')+
scale_x_log10(limits = c(0.8,10), breaks = seq(0.8,10,1))+
facet_wrap( ~ quality,ncol=2, scales = 'free_y')+
labs(title = 'Faceting the distribution of residual sugar content based on wine quality in the available red wines')
The above diagram shows that almost all qualities of red wines use residual sugar in amounts between 1.5 to 2.5 grams/liter. And scarcely do the values of the residual sugar level varies. So chances are that the presence of residual sugar must be almost fixed in all red wines.
## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.200 1.875 2.100 2.635 3.100 5.700
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.300 1.900 2.100 2.694 2.800 12.900
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.200 1.900 2.200 2.529 2.600 15.500
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.900 1.900 2.200 2.477 2.500 15.400
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.200 2.000 2.300 2.721 2.750 8.900
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.400 1.800 2.100 2.578 2.600 6.400The above statistics show the summary of residual sugar present in different qualities of red wine.
ggplot(wine_df, aes(x = quality, y = residual.sugar))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(1.5,3.3))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title = 'Boxplot of amount of residual sugar level vs Quality of red wines')
The above graph using boxplots and the statistical summary above that proves that in almost all qualities of red wines the median quantity of residual sugar ranges from 2.1 to 2.3 grams/liter. And the mean quantity of residual sugar is around 2.5 grams/liter. The red point shows the mean value of the residual sugar.
The huge variation between mean and median is due to the presence of outliers.
Analyzing variable : chlorides
Description of chlorides: The amount of salt in the wine.
ggplot(wine_df, aes(x=chlorides))+
geom_histogram(binwidth = 0.025,color = 'black', fill='white')+
scale_x_continuous(limits = c(0,0.62), breaks = seq(0,0.62,0.05))+
labs(title = 'Histogram of amount of chlorides in the available red wines')
The above histogram has a right skewed distribution. And it can be concluded from the distribution that chlorides added to the red wines are in the range of 0.05 to 0.1 grams.
ggplot(wine_df, aes(x=chlorides))+
geom_histogram(binwidth = 0.025,color = 'black', fill='white')+
scale_x_continuous(limits = c(0,0.62), breaks = seq(0,0.62,0.05))+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')+
labs(title = 'Faceting the distribution of chlorides based on wine quality in the available red wines')
The above faceted graph shows almost all qualities of wine have their chlorides amount in the range of 0.05 to 1 grams. So it can be inferred that it is necessary to have chlorides in the abive mentioned range for the preapartaion of the above wine.
by(wine_df$chlorides, wine_df$quality, summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0610 0.0790 0.0905 0.1225 0.1430 0.2670
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.04500 0.06700 0.08000 0.09068 0.08900 0.61000
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.03900 0.07400 0.08100 0.09274 0.09400 0.61100
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.03400 0.06825 0.07800 0.08496 0.08800 0.41500
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.01200 0.06200 0.07300 0.07659 0.08700 0.35800
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.04400 0.06200 0.07050 0.06844 0.07550 0.08600ggplot(wine_df, aes(x=quality,y=chlorides))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(0.05,0.14))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title='Boxplot of amount of chlorides vs the quality of red wine')
Using the boxplot and statistical summary it is seen that chlorides are used in small quanities of 0.05 to 0.15 grams. And the best available red wine of quality 8 uses the least median amount of 0.0705 grams and as the quality of red wine decreases the chloride content increases along with it. The red point shows the mean value of the chlorides.
The huge difference between mean and median is due to the presence of these outliers.
Analyzing variable : free.sulfur.dioxide
Description of free.sulfur.dioxide: The free form of SO2 exists in equilibrium between molecular SO2 (as a dissolved gas) and bisulfite ion; it prevents microbial growth and the oxidation of wine
ggplot(wine_df, aes(x=free.sulfur.dioxide))+
geom_histogram(binwidth = 1, color = 'black', fill='white')+
scale_x_continuous(limits = c(1,72), breaks = seq(1,72,2))+
labs(title = 'Histogram of free.sulfur.dioxide content in the available red wines')
The distribution of free.sulfur.dioxide content in the available red wines is right-skewed in nature and its mode value at 5.
ggplot(wine_df, aes(x=free.sulfur.dioxide))+
geom_histogram(binwidth = 2, color = 'black', fill='white')+
scale_x_continuous(limits = c(1,72), breaks = seq(1,72,4))+
facet_wrap( ~ quality, ncol = 2, scales='free_y') +
labs(title = 'Faceting the distribution of free.sulfur.dioxide based on wine quality')
The faceted distribution of free sulfur dioxide based on quality shows that majority of almost all qualities of wine use free sulifur dioxide in 5 ppm amounts.
by(wine_df$free.sulfur.dioxide, wine_df$quality, summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.0 5.0 6.0 11.0 14.5 34.0
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.00 6.00 11.00 12.26 15.00 41.00
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.00 9.00 15.00 16.98 23.00 68.00
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1.00 8.00 14.00 15.71 21.00 72.00
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.00 6.00 11.00 14.05 18.00 54.00
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.00 6.00 7.50 13.28 16.50 42.00ggplot(wine_df, aes(x=quality,y=free.sulfur.dioxide))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(0,25))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title='Boxplot of amount of free.sulfur.dioxide vs the quality of red wine')
Using the boxplot and statistical summary it can be inferred that even though the presence of free sulfur dioxide is essential in red quality wines to prevent the growth of microbes, its amount in the wine does not have much affect on its quality. The red point shows the mean value of the free sulfur dioxide.
Analyzing variable : total.sulfur.dioxide
Description of total.sulfur.dioxide: Amount of free and bound forms of S02 in low concentrations. S02 is mostly undetectable in wine but at free S02 concentrations over 50 ppm, S02 becomes evident in the nose and taste of wine.
ggplot(wine_df, aes(x=total.sulfur.dioxide))+
geom_histogram(binwidth = 10,fill='white',color='black')+
scale_x_continuous(limits = c(0,300), breaks = seq(0,300,10))+
labs(title = 'Histogram of distribution of sulfur.dioxide content in the available red wines')
The given histogram is right skewed in nature. And most of the available red quality wines have total sulfur dioxide of amount about 20 to 30 ppm.
ggplot(wine_df, aes(x=total.sulfur.dioxide))+
geom_histogram(binwidth = 10,fill='white',color='black')+
scale_x_continuous(limits = c(0,180), breaks = seq(0,180,10))+
facet_wrap( ~ quality, ncol = 2, scales='free_y') +
labs(title = 'Faceting the distribution of total.sulfur.dioxide based on wine quality')
The above faceted graph shows that majority of all qualities of wine have total sulfur dioxide in the of 20 to 30 ppm. And the histogram corresponding to each quality has a right-skewed distribution.
by(wine_df$total.sulfur.dioxide, wine_df$quality,summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 9.0 12.5 15.0 24.9 42.5 49.0
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 7.00 14.00 26.00 36.25 49.00 119.00
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 6.00 26.00 47.00 56.51 84.00 155.00
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 6.00 23.00 35.00 40.87 54.00 165.00
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 7.00 17.50 27.00 35.02 43.00 289.00
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 12.00 16.00 21.50 33.44 43.00 88.00ggplot(wine_df, aes(x=quality,y=total.sulfur.dioxide))+
geom_boxplot(aes(group=quality))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title='Boxplot of amount of total.sulfur.dioxide vs the quality of red wine')+
coord_cartesian(ylim = c(5,90))
From the statistical summary and boxplot it is seen that only quality 5 red wines have the highest content of total sulfur dioxide with median value of 47 ppm and mean value of 56.51 ppm. Also the best quality of available red wine which is quality 8 have low median content of total sulfur dioxide with 21.5 ppm and low mean content of total sulfur dioxide 33.44. And quality 3 which has low median content of 15 ppm and mean content of 24.9 ppm. So based on the amount of total sulfur dioxide available in each quality of wine we cannot decide whether quality of red wine is good or not. The red point shows the mean value of the total sulfur dioxide.
Analyzing variable : density
Description of density: the density of water is close to that of water depending on the percent alcohol and sugar content.
ggplot(wine_df, aes(x=density))+
geom_histogram(binwidth = 0.001, color = 'black', fill = 'white')+
labs(title='Histogram of density of red wines')+
scale_x_continuous(breaks = seq(0.990,1.004,0.001), limits = c(0.990,1.004) )
The above histogram on density of red wines has a normal distribution and most of the available red wines have a density in the range of 0.995 to 0.997.
ggplot(wine_df, aes(x=density))+
geom_histogram(binwidth = 0.001, color = 'black', fill = 'white')+
labs(title='Faceting the distribution of density based on wine quality of the available red wines')+
scale_x_continuous(breaks = seq(0.990,1.004,0.002), limits = c(0.990,1.004))+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')
The faceted distribution shows that almost all qaulity of red wines have a normal distribution with their mean value of density concentrated around the 0.994 to 0.996 region of density for quality 7 and quality 8 red wines. And the other lower quality wines had their majority of the density values ranging from 0.995 to 0.997.
by(wine_df$density, wine_df$quality, summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.9947 0.9961 0.9976 0.9975 0.9988 1.0008
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.9934 0.9957 0.9965 0.9965 0.9974 1.0010
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.9926 0.9962 0.9970 0.9971 0.9979 1.0031
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.9901 0.9954 0.9966 0.9966 0.9979 1.0037
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.9906 0.9948 0.9958 0.9961 0.9974 1.0032
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.9908 0.9942 0.9949 0.9952 0.9972 0.9988ggplot(wine_df, aes(x=quality, y=density))+
geom_boxplot(aes(group=quality))+
labs(title='Boxplot of density vs quality of red wine')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
coord_cartesian(ylim = c(0.994,0.999))
From the statistical summary and boxplot it can be seen that good quality red wines having quality 8 and 7 have a low median value of density 0.9949 and 0.9958 respectively, and mean density value of 0.9952 and 0.9961. For other lower qualities of wine such as quality 3 has median density value of 0.9976 and mean density value of 0.9975. So it can be said that lower density of red wines denotes better quality. The red point shows the mean value of the density.
Analyzing variable : pH
Description of pH: describes how acidic or basic a wine is on a scale from 0 (very acidic) to 14 (very basic); most wines are between 3-4 on the pH scale.
ggplot(wine_df, aes(x=pH))+
geom_histogram(binwidth = 0.1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(2.73, 4.02), breaks = seq(2.74,4.02,0.1))+
labs(title = 'Histogram of pH values of red wine')
The above histogram has a normal distribution and most of the red wines have pH values in the range of 3.14 to 3.44.
ggplot(wine_df, aes(x=pH))+
geom_histogram(binwidth = 0.1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(2.73, 4.02), breaks = seq(2.74,4.02,0.1))+
labs(title = 'Faceting the distribution of pH values based on wine quality in the available red wines')+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')
From the above faceted graph it can be seen that quality 5 to 8 have normal distribution and their values range from 3.14 to 3.44. The graph of distribution of pH values for different qualities of red wine are not much helpful in finding how pH values affect quality.
by(wine_df$pH, wine_df$quality,summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 3.160 3.312 3.390 3.398 3.495 3.630
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.740 3.300 3.370 3.382 3.500 3.900
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.880 3.200 3.300 3.305 3.400 3.740
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.860 3.220 3.320 3.318 3.410 4.010
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.920 3.200 3.280 3.291 3.380 3.780
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 2.880 3.163 3.230 3.267 3.350 3.720ggplot(wine_df, aes(x=quality,y=pH))+
geom_boxplot(aes(group=quality))+
labs(title = 'Boxplot of pH vs Quality for red wine')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
coord_cartesian(ylim = c(3.15,3.55))
From the statistical summary and boxplot it is observed that median pH value of quality 8 red wine is 3.23 and it is lower than other qualities of red wine but its mean pH value is 3.267 which is higher than the mean pH values of other quality of wines. So it cannot be generally stated that pH has a direct relationship with the quality of wine. The red point shows the mean value of the pH.
Analyzing variable : sulphates
Description of sulphates: a wine additive which can contribute to sulfur dioxide gas (S02) levels, wich acts as an antimicrobial and antioxidant.
ggplot(wine_df, aes(x=sulphates))+
geom_histogram(binwidth = 0.1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(0.3,2), breaks = seq(0.3,2,0.1))+
labs(title = 'Histogram of distribution of sulphates for red wines')
The distribution of sulphates in red wine has a right-skewed distribution with the mode value of sulphates lying in the range from 0.5 to 0.6.
ggplot(wine_df, aes(x=sulphates))+
geom_histogram(binwidth = 0.1, color = 'black', fill = 'white')+
scale_x_continuous(limits = c(0.3,2), breaks = seq(0.3,2,0.1))+
labs(title = 'Faceting the distribution of sulphates based on wine quality in the available red wines')+
facet_wrap( ~ quality, ncol = 2, scales = 'free_y')
From the faceted distribution of sulphates considering quality as the categorical variable it is seen that that the amount of sulphates in higher quality of wines such as quality 8 and 7 have slightly higher higher amount of sulphates than those ranging from quality 3 to quality 6.
by(wine_df$sulphates, wine_df$quality,summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.4000 0.5125 0.5450 0.5700 0.6150 0.8600
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.3300 0.4900 0.5600 0.5964 0.6000 2.0000
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.370 0.530 0.580 0.621 0.660 1.980
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.4000 0.5800 0.6400 0.6753 0.7500 1.9500
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.3900 0.6500 0.7400 0.7413 0.8300 1.3600
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.6300 0.6900 0.7400 0.7678 0.8200 1.1000ggplot(wine_df, aes(x=quality,y=sulphates))+
geom_boxplot(aes(group=quality))+
labs(title='Boxplot of amount of sulphates vs Quality for red wine')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4) +
coord_cartesian(ylim = c(0.48,0.84))
From the statistical summary and boxplot of sulphates vs the quality of red wines it is observed that better quality red wines such as quality 8 and 7 have higher median amd mean value of sulphates as compared to that of other lower quality red wines. The red point shows the mean value of the sulphates.
Analyzing variable : alcohol
Description of alcohol: the percent alcohol content of the wine measured in % by volume.
ggplot(wine_df, aes(x=alcohol))+
geom_histogram(binwidth = 0.5,color='black',fill='white')+
labs(title = 'Histogram of distribution of alcohol in red wines')+
scale_x_continuous(limits = c(8,15), breaks = seq(8,15,0.5))
The alcohol content in red wines have a right-skewed distribution with most of the red wines having alcohol content in range from 9-10 % by volume.
ggplot(wine_df, aes(x=alcohol))+
geom_histogram(binwidth = 0.5,color='black',fill='white')+
labs(title = 'Faceting the distribution of alcohol based on wine quality in the available red wines')+
scale_x_continuous(limits = c(8,15), breaks = seq(8,15,0.5))+
facet_wrap( ~quality, ncol=2, scales = 'free_y')
From the faceted distribution of alcohol based on quality it can be observed that usually good quality red wines having quality ratings of quality 8 and 7 have higher alcohol content than other quality of available red wines.
by(wine_df$alcohol, wine_df$quality, summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 8.400 9.725 9.925 9.955 10.575 11.000
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 9.00 9.60 10.00 10.27 11.00 13.10
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 8.5 9.4 9.7 9.9 10.2 14.9
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 8.40 9.80 10.50 10.63 11.30 14.00
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 9.20 10.80 11.50 11.47 12.10 14.00
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 9.80 11.32 12.15 12.09 12.88 14.00ggplot(wine_df, aes(x=quality,y=alcohol))+
geom_boxplot(aes(group=quality))+
labs(title='Boxplot of alcohol content vs quality of red wine')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
coord_cartesian(ylim = c(9.2,13))
From the statistical summary and boxplot above it can be assumed that high quality of red wines have higher alcohol content % by volume as compared to other lower quality wines. The red point shows the mean value of the alcohol content in percentage by volume.
Analyzing created variable : Total_acidity
Description of Total_acidity : As there are three types of acidity in the red wine so i was just curious to find out the net effect of all acidity on the quality of red wine.
So i created a new variable called Total_acidity (measured in g/dm3) which is the sum of fixed acidity, volatile acidity and citirc acid.
wine_df <- wine_df %>%
mutate(Total_acidity = fixed.acidity + citric.acid + volatile.acidity)
ggplot(wine_df, aes(x=Total_acidity))+
geom_histogram(binwidth = 0.5, color='black', fill='white')+
scale_x_continuous(limits=c(5,17.5), breaks = seq(5,17.5,0.5))+
labs(title='Histogram of distribution of total_acidity in red wines')
The above histogram has a right-skewed distribution with majority of the red wines having total acidity in the range of 7.5 to 9 g/dm3.
ggplot(wine_df, aes(x=Total_acidity))+
geom_histogram(binwidth = 0.5, color='black', fill='white')+
scale_x_continuous(limits=c(5,17.5), breaks = seq(5,17.5,1))+
labs(title='Faceting the distribution of total_acidity in red wines based on quality')+
facet_wrap( ~quality, ncol=2, scales = 'free_y')
The faceted graph shows the distribution of total acidity in the available red wine for different qualities.Quality 6 and 5 have a right-skewed distribution. And the distibution for better qualilty 7 and 8 have uneven distribution.
by(wine_df$Total_acidity, wine_df$quality,summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 7.480 8.101 8.883 9.415 10.780 12.840
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.270 7.570 8.300 8.647 9.300 13.450
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.770 7.880 8.600 8.988 9.830 16.910
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.390 7.710 8.640 9.118 10.186 15.350
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.320 8.105 9.470 9.651 10.980 17.045
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 5.660 7.933 9.095 9.381 11.002 13.630ggplot(wine_df, aes(x=quality, y=Total_acidity))+
geom_boxplot(aes(group=quality))+
labs(title='Boxplot of total acidity vs quality of red wine')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
coord_cartesian(ylim = c(7.5,11))
The statistical summary and boxplot of total acidity with respect to quality does not show any type of relation between them. The red point shows the mean value of the total acidity.
Analyzing created variable : prop_free_so2
Description of prop_free_so2 : It is the ratio of free.sulfur.dioxide to total.sulfur.dioxide present in the concerned red wine. As the total.sulfur.dioxide present in the red wine contains both free.sulfur.dioxide and bound.sulfur.dioxide so to determine the proportion in which the free.sulfur.dioxide is used and to check how this ratio affect the quality of red wine we use this variable.
wine_df <- wine_df %>%
mutate(prop_free_so2 = free.sulfur.dioxide/total.sulfur.dioxide)
ggplot(wine_df, aes(x=prop_free_so2))+
geom_histogram(binwidth = 0.1, color='black',fill='white')+
labs(title='Histogram of distribution of proportion of free SO2 in red wine')+
scale_x_continuous(limits = c(0,1), breaks = seq(0,1,0.1))
The above histogram shows distribution of proportion of free SO2 in the red wine and most of the red wines have the proportion usually between 0.1 to 0.5.
ggplot(wine_df, aes(x=prop_free_so2))+
geom_histogram(binwidth = 0.1, color='black',fill='white')+
labs(title='Faceting the distribution of proportion of free SO2 in red wine based on quality')+
scale_x_continuous(limits = c(0,1), breaks = seq(0,1,0.1))+
facet_wrap( ~ quality, ncol=2, scales = 'free_y')
The above faceted distribution shows distribution of proportion of free SO2 in various qualities of red wine.
by(wine_df$prop_free_so2, wine_df$quality,summary)## wine_df$quality: 3
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.2128 0.3237 0.4541 0.4366 0.5502 0.7083
## --------------------------------------------------------
## wine_df$quality: 4
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0531 0.2727 0.3913 0.3875 0.5000 0.7083
## --------------------------------------------------------
## wine_df$quality: 5
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.08065 0.23077 0.32394 0.33917 0.42857 0.78261
## --------------------------------------------------------
## wine_df$quality: 6
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.02273 0.28571 0.40362 0.40962 0.50000 0.85714
## --------------------------------------------------------
## wine_df$quality: 7
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.1298 0.3333 0.4500 0.4361 0.5357 0.7429
## --------------------------------------------------------
## wine_df$quality: 8
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.1364 0.3133 0.3823 0.4065 0.4926 0.7556ggplot(wine_df, aes(x=quality, y=prop_free_so2))+
geom_boxplot(aes(group=quality))+
coord_cartesian(ylim = c(0.2,0.6))+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4) +
labs(title='Boxplot of free sulfur dioxide to the quality of red wine')
The statistical summary and boxplot gives the median and mean value of proportion of free SO2 with respect to total sulfur dioxide. The red point shows the mean value of the proportion of free sulfur dioxide.
Univariate Analysis
What is the structure of your dataset?
The above dataset consists of 1599 observations and 15 variables.
Also i added 2 variables to the dataset namely Total_acidity and prop_free_SO2.
What is/are the main feature(s) of interest in your dataset?
In the given dataset, 6 qualities of datset are available namely Quality 3, Quality 4, Quality 5, Quality 6, Quality 7 and Quality 8.
Of the available qualities of red wines quality 5 and quality 6 are the majority 681 and 638 respectively.
What other features in the dataset do you think will help support your
investigation into your feature(s) of interest?
The following analysis made in the above univariate analysis might help in determining the chemical property which help in influencing the quality of red wines-
- Good quality red wines have lower amounts of volatile acids present in them.
- Citric acid content is higher in good quality wines.
- Amount of residual sugar content is almost same across all qualities of red wine.
- Good quality of red wines have lower density.
- Sulphate content is higher in good quality red wines.
- Alcohol content is higher in good quality red wines.
Did you create any new variables from existing variables in the dataset?
Yes, i created 2 new variables from the existing variables.
Total_acidity : As there are three types of acidity in the red wine so i was just curious to find out the net effect of all acidity on the quality of red wine. So i created this variable which is the sum of fixed
acidity, volatile acidity and citirc acid. (measured in g/dm3)prop_free_so2 : It is the ratio of free.sulfur.dioxide to total.sulfur.dioxide present in the concerned red wine. As the total.sulfur.dioxide present in the red wine contains both free.sulfur.dioxide and bound.sulfur.dioxide so to determine the proportion in which the free.sulfur.dioxide is used and to check how this ratio affect the quality of red wine we use this variable.
Of the features you investigated, were there any unusual distributions?
Did you perform any operation on the data to tidy, adjust, or change the form
of the data? If so, why did you do this?
I tried to change the scale of x-axis from continuous to logarithmic but there was not much variation to the plot. No other transformation to the dataset and plot was required as
Bivariate Plots Section
In the bivariate analysis, i will start by finding the correlation coefficent between quality and other variables and among the other supporting variables.
Correlation matrix
cwine_df <- wine_df %>%
select(c(2:15))
mcor <- cor(cwine_df)
round(mcor,digits = 3) ## fixed.acidity volatile.acidity citric.acid
## fixed.acidity 1.000 -0.256 0.672
## volatile.acidity -0.256 1.000 -0.552
## citric.acid 0.672 -0.552 1.000
## residual.sugar 0.115 0.002 0.144
## chlorides 0.094 0.061 0.204
## free.sulfur.dioxide -0.154 -0.011 -0.061
## total.sulfur.dioxide -0.113 0.076 0.036
## density 0.668 0.022 0.365
## pH -0.683 0.235 -0.542
## sulphates 0.183 -0.261 0.313
## alcohol -0.062 -0.202 0.110
## quality 0.124 -0.391 0.226
## Total_acidity 0.996 -0.204 0.690
## prop_free_so2 -0.131 -0.073 -0.167
## residual.sugar chlorides free.sulfur.dioxide
## fixed.acidity 0.115 0.094 -0.154
## volatile.acidity 0.002 0.061 -0.011
## citric.acid 0.144 0.204 -0.061
## residual.sugar 1.000 0.056 0.187
## chlorides 0.056 1.000 0.006
## free.sulfur.dioxide 0.187 0.006 1.000
## total.sulfur.dioxide 0.203 0.047 0.668
## density 0.355 0.201 -0.022
## pH -0.086 -0.265 0.070
## sulphates 0.006 0.371 0.052
## alcohol 0.042 -0.221 -0.069
## quality 0.014 -0.129 -0.051
## Total_acidity 0.124 0.117 -0.154
## prop_free_so2 -0.071 -0.105 0.327
## total.sulfur.dioxide density pH sulphates alcohol
## fixed.acidity -0.113 0.668 -0.683 0.183 -0.062
## volatile.acidity 0.076 0.022 0.235 -0.261 -0.202
## citric.acid 0.036 0.365 -0.542 0.313 0.110
## residual.sugar 0.203 0.355 -0.086 0.006 0.042
## chlorides 0.047 0.201 -0.265 0.371 -0.221
## free.sulfur.dioxide 0.668 -0.022 0.070 0.052 -0.069
## total.sulfur.dioxide 1.000 0.071 -0.066 0.043 -0.206
## density 0.071 1.000 -0.342 0.149 -0.496
## pH -0.066 -0.342 1.000 -0.197 0.206
## sulphates 0.043 0.149 -0.197 1.000 0.094
## alcohol -0.206 -0.496 0.206 0.094 1.000
## quality -0.185 -0.175 -0.058 0.251 0.476
## Total_acidity -0.096 0.676 -0.683 0.182 -0.067
## prop_free_so2 -0.371 -0.265 0.185 -0.010 0.246
## quality Total_acidity prop_free_so2
## fixed.acidity 0.124 0.996 -0.131
## volatile.acidity -0.391 -0.204 -0.073
## citric.acid 0.226 0.690 -0.167
## residual.sugar 0.014 0.124 -0.071
## chlorides -0.129 0.117 -0.105
## free.sulfur.dioxide -0.051 -0.154 0.327
## total.sulfur.dioxide -0.185 -0.096 -0.371
## density -0.175 0.676 -0.265
## pH -0.058 -0.683 0.185
## sulphates 0.251 0.182 -0.010
## alcohol 0.476 -0.067 0.246
## quality 1.000 0.104 0.194
## Total_acidity 0.104 1.000 -0.149
## prop_free_so2 0.194 -0.149 1.000Plotting the Correlation matrix
corrplot(mcor)
From the correlation matrix plot it can be observed that :
- Fixed acidity is highly correlated citric acid, density and Total acidity. Fixed acidity has high value of negative correlation with pH. And correlation of fixed acidity with quality is very low.
- Volatile acidity is negatively correlated with citric acid and quality.
- Citric acid has high positive correlation with fixed acidity and total acidity.
- Density has highly positively correlated with fixed acidity and total acidity.
- pH has negative correlation with density, fixed, citric and total acidity.
- Quality has negative correlation with density and positive correlation with alcohol.
Scatterplot matrix
splom(cwine_df[c(1,2,7,8,9,11,12)], data = cwine_df, groups = NULL, axis.line.tck=0,axis.text.alpha = 0)
cwine_df <- cwine_df %>%
select(1,2,7,8,9,11,12)The above scatter plot matrix has been made by selecting the variables from the correlation matrix plot. This graph helps in getting the idea of scatterplot that we will get when plotting different variables in the datset,
p2 <- ggpairs(data = cwine_df)+
labs(title='Generalized Pairs Plot')
p2
The generalized pairs plot shown above shows the correlation and scatterplot between selected pairs of variables from the correlation and scatterplot matrix shown above. The correlation values between variables are shown in the upper triangular matrix. The diagonal matrix has the pairing between same variables. The lower triangular shows the scatterplot between the pairs of variables. The correlation values and scatterplot between variables helps in understanding the relationship between the variables.
ggplot(wine_df, aes(x=pH, y=fixed.acidity))+
geom_point(color='grey60')+
stat_smooth(color='red',method='lm')+
labs(title='Scatterplot of Fixed-acidity vs pH of red wine',y='fixed acidity (tartaric acid-g/dm^3)')
The above graph shows the negative correlation between fixed acidity and pH value of red wine. Also a linear regression line is added to a scatterplot.
ggplot(wine_df, aes(x=pH, y=density))+
geom_point(color='grey60')+
stat_smooth(color='red')+
labs(title='Scatterplot of density vs pH of red wine',y='density (g/cm^3)')
The above graph shows the scatterplot between density and pH of red wine it shows the negative corelation between density and pH. Also the regression line used here is a loess curve.
ggplot(wine_df,aes(y=density,x=alcohol))+
geom_point(color='grey60')+
stat_smooth(method = 'loess',color='red', linetype='solid')+
labs(title='Scatterplot of density vs alcohol content in red wine',
x = 'alcohol (% by volume)', y='density (g/cm^3)')
The above graph shows scatterplot of density vs alcohol content in red wine. These parameters are negatively correlated and the negative correlation is shown using a loess curve.
ggplot(wine_df,aes(x=factor(quality), y = alcohol))+
geom_jitter(alpha=0.3)+
geom_boxplot(alpha=0.8, color = 'blue')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title='Scatterplot and Boxplot of Quality vs alcohol content of red wine',
y = 'alcohol (% by volume)')+
coord_cartesian(ylim = c(9,13))
The above scatterplot has quality as discrete variable and for higher quality of red wine the alcohol content is also high which also proves the positive correlation between them. The blue line denotes the mean value of alcohol content across each quality. The red line denotes the median value of alcohol content across each quality.
In the univariate analysis, boxplot graph was shown to prove the relation between quality and alcohol content in red wine.
ggplot(wine_df, aes(x=factor(quality), y=volatile.acidity))+
geom_jitter(alpha=1/3)+
geom_boxplot(alpha=0.8, color = 'blue')+
stat_summary(fun.y = 'mean', geom = 'point',color='red',shape=8,size=4)+
labs(title='Scatterplot of Quality vs volatile acidity of red wine',
y='volatile acidity (acetic acid - g/dm^3)')+
coord_cartesian(ylim = c(0.2,1.1))
The above scatterplot has quality as discrete variable and for higher quality of red wine the volatile acidity is very low which also proves the negative correlation between them. The blue line denotes the mean value of volatile acidity across each quality. The red line denotes the median value of volatile acidity across each quality.
In the univariate analysis, boxplot graph was shown to prove the relation between quality and volatile acidity in red wine.
Bivariate Analysis
Talk about some of the relationships you observed in this part of the
investigation. How did the feature(s) of interest vary with other features in
the dataset?
The bivarite analysis gave an insight into the correlation between quality and other variables with the help of correlation coefficient and scatterplot.It was observed that quality of red wine improves with the decreases in volatile acidity. The alcohol content in red wine also improves the wine quality. From the correlation matrix it was also observed that residual sugar and other parameters don’t have much effect on the quality of red wine.
Did you observe any interesting relationships between the other features
(not the main feature(s) of interest)?
The relationship between pH and fixed/total acidity they have negative correlation i.e., as pH value increases acidity content decreases. Also, pH is negatively correlated with the density of the red wine. And density of the red wine is also having negative correlation with the alcohol content in the red wine. It is also found that citric acid is positively correlated with fixed acidity of red wine but has a negative correlation with volatile acidity.
The bivariate analysis helped in improving the insight into the relationship between quality and variables such as residual sugar,density and citric acid.It was found that the presence of these variables don’t have much impact on the quality of red wines.
What was the strongest relationship you found?
The quality of red wine is highly dependant on volatile acidity and alcohol content of the red wine. The quality and volatile acidity has negative correlation. The quality and alcohol content has a positive correlation between them.
Multivariate Plots Section
# Converting quality to ordered factors type
wine_df$quality <- ordered(wine_df$quality, levels=c(1:10))
ggplot(data = wine_df,aes(y=density, x=alcohol,color=quality))+
geom_point(size=1,alpha=0.8)+
geom_smooth(method = 'lm', se = FALSE, size=2)+
scale_color_brewer(type='seq', guide = guide_legend(title = 'Quality'))+
ggtitle('Alcohol and density of red wine based on Quality')+
labs(x = 'Alcohol (% by volume)',y='density (g/cm^3)')
Analyzing the above plot it’s seen that as the quality of red wine improves it’s density decreases and alcohol content also increases as the points seem to move towards the bottom right corner of the graph.
p3 <- ggplot(data = wine_df,aes(x=pH, y=fixed.acidity,color=quality))+
geom_point(size=1,alpha=0.8)+
geom_smooth(method = 'lm', se = FALSE, size=2)+
scale_color_brewer(type='seq', guide = guide_legend(title = 'Quality'))+
ggtitle('pH and fixed acidity of red wine based on Quality')+
labs(y='fixed acidity (tartaric acid - g/dm^3)')
p3
It is observed from the scatterplot that the pH value of wine lies between 3 and 4 . Also for a good quality of red wine pH value must be less than 3.5 as most points are concentrated there. As it is observed from the linear regression lines as pH decreases fixed acidity content also improves.
I also referred the following blog to know more about the pH value of red wines. Blog on pH of wines
p3 <- ggplot(wine_df, aes(x=alcohol,y=volatile.acidity,color=quality))+
geom_point(size=1,alpha=0.8)+
geom_smooth(method = 'lm', se = FALSE, size=2)+
scale_color_brewer(type='seq', guide = guide_legend(title = 'Quality'))+
ggtitle('volatile acidity vs alcohol content of red wine based on quality')+
labs(x = 'Alcohol (% by volume)', y = 'volatile acidity (acetic acid - g/dm^3)')Alcohol content and volatile acidity of wine are the two important factors that affect the quality of red wine as from all the analysis done above. It is seen that as volatile acidity decreases and alcohol content improves the quality of wine improves gradually. So referring to the graph, as amount of volatile acidity the quality of wine also improves but the only variation which is observed is in case of Quality 8 red wine. It can be observed that the slope of other quality of red wines is almost zero but for quality 8 red wine it can be seen that slope has a low positive value. So a slight increase in the volatile acidity content is marked by high increase in the alcohol content to maintain the quality of red wine.
Multivariate Analysis
Talk about some of the relationships you observed in this part of the
investigation. Were there features that strengthened each other in terms of
looking at your feature(s) of interest?
The relation between fixed acidity and pH by quality strengthens each other. As its known that low pH value implies high acidity so from the graph it is seen that at low pH values fixed acidity value is high. ANd the quality of wine also improves as the above relationship in the wine is satisfied.
Were there any interesting or surprising interactions between features?
One surprising feature is the relationship between density and alcohol content by quality. It is seen that as the alcohol content decreases the density of the red wine has increased and the quality decreases that is the good quality wines points tend to move towards the right bottom corner of the graph with low density and high alcohol content.
Final Plots and Summary
Plot One
p1
Description One
It is observed that wines with quality ratings 5and 6 is the most available in the given dataset. Followed by quality 7 and quality 4. Quality 8 followed by quality 3 are the least available. Data below gives the count of each quality of red wine available in the given dataset.
table(wine_df$quality)##
## 1 2 3 4 5 6 7 8 9 10
## 0 0 10 53 681 638 199 18 0 0Plot Two
Plotting the Correlation matrix
p2
Description Two
The generalized pairs plot shown above shows the correlation and scatterplot between selected pairs of variables from the correlation and scatterplot matrix shown above. The correlation values between variables are shown in the upper triangular matrix. The diagonal matrix has the pairing between same variables. The lower triangular shows the scatterplot between the pairs of variables. The correlation values and scatterplot between variables helps in understanding the relationship between the variables.
It can be observed that :
- Fixed acidity is highly correlated citric acid, density and Total acidity. Fixed acidity has high value of negative correlation with pH. And correlation of fixed acidity with quality is very low.
- Volatile acidity is negatively correlated with citric acid and quality.
- Citric acid has high positive correlation with fixed acidity and total acidity.
- Density has highly positively correlated with fixed acidity and total acidity.
- pH has negative correlation with density, fixed, citric and total acidity.
- Alcohol has negative correlation with density and positive correlation with alcohol.
Plot Three
p3
Description Three
Alcohol content and volatile acidity of wine are the two important factors that affect the quality of red wine based on correlation values. It is seen that as volatile acidity decreases and alcohol content improves the quality of wine improves gradually. So referring to the graph, as amount of volatile acidity the quality of wine also improves but the only variation which is observed is in case of Quality 8 red wine. It can be observed that the slope of other quality of red wines is almost zero but for quality 8 red wine it can be seen that slope has a low positive value. So a slight increase in the volatile acidity content is marked by high increase in the alcohol content to maintain the quality of red wine.
Reflection
The conclusions drawn by me in this analysis are:
- In the given dataset for red wine, Quality 5 and quality 6 wines are the most available.
- Red wines of Quality 1, Quality 2, Quality 9 and Quality 10 are not available.
- Amount of residual sugar levels (in g/dm3) in red wines have no affect on quality of red wine,
- As pH decreases fixed acidity amount improves.
- Density of red wine decreases montonically as pH value increases.
- Density of red wine decreases with increase in alcohol content.
- Quality of red wine improves as alcohol content increases.
- Quality of red wine improves as volatile acid’s content decreases.
The analysis could have been performed better if there was more data on other qualities of red wines available. If more data had been availabe results could have been more accurate and new conclusions and inferences could have been made.
It was surprising to see that some chemical parameters such as residual sugar, chlorides, free sulfur dioxide and total sulfur dioxide did not have much affect on the quality of red wine.
For future explorations i can combine the red wine dataset with the dataset for white wine to draw comparisons between them and analyze the differences in their chemical constituents or parameters. Also the data for other qualities of red wine would greatly help in improving this analysis.