Wine quality Analysis

WINE QUALITYANALYSIS

ANKIT HALDARASHITHA VSDEBARNIK BISWASKRISHNA BOLLOJULA

AGENDA• OBJECTIVE• DATA DESCRIPTION• DATA ANALYSIS• INSIGHTS FROM ANALYSIS

OBJECTIVE• The dataset contains information about red

and white wine.• The dataset has 1000 data points of each

wine with 13 characteristics(attributes).• Wine quality is measured on 0(low)-

10(high) scale.• Management want

– To understand the characteristics of these wines.

– How different ingredients affect the quality.

DATA DESCRIPTION• The attributes are as follows

• fixed.acidity• volatile.acidity• citric.acid• residual.sugar• chlorides• free.sulfur.dioxide

• total.sulfur.dioxide

• density• pH• sulphates• alcohol• quality• wine_type

DATA ANALYSIS• To understand the characteristics of the

wine and find ingredients contributing more for quality, we have done the following– Subset the data based on wine type and find

descriptive statistics.– Subset the data based on good quality(7,8,9)

and bad quality(3,4,5,6) and find descriptive statistics.

– Decision tree.– Clustering the dataset.

DATA ANALYSIS• Subset the data based on wine type and find descriptive statistics.

• Subset the data based on good quality1(7,8,9) and bad quality0(3,4,5,6) and found descriptive statistics.

DATA ANALYSIS• Subset the data based on wine quality and find descriptive statistics.

• Volatile.acidity, citric.acid, chlorides are showing some pattern. Alcohol being the best.

DATA ANALYSIS• Decision tree

– Using rapid miner, we build a classification model to find the ingredients which are important for predicting red wine and white wine.

accuracy: 97.05% +/- 1.63% (mikro: 97.05%)

true Red Wine true White Wine class precision

pred. Red Wine 961 20 97.96%

pred. White Wine 39 980 96.17%

class recall 96.10% 98.00%

Ingredientschlorides

sulphatesfree.sulfur.dioxidevolatile.aciditytotal.sulfur.dioxide

DATA ANALYSIS• Decision tree

– Using rapid miner, we build a classification model to find the ingredients which are important for predicting the quality is high or low.

Ingredientscitric.acid

chloridesvolatile.aciditydensityalcohol

true 0 true 1 class precision

pred. 0 1431 183 88.66%pred. 1 211 175 45.34%class recall 87.15% 48.88%

DATA ANALYSIS• Clustering(Code is on next slide)

• The cluster divides the whole data based on which wine it belongs.• We can see there is significant difference in all attributes.

DATA ANALYSIS

INSIGHTS FROM ANALYSIS

• Characteristics– Red wine or white wine– High or low quality

• Ingredients which contribute more for type of the wine.– Chlorides, sulphates, free.sulfur.dioxide, total

sulfur dioxide, volatile.acidity.• Ingredients which contribute more for

quality of the wine.– Alcohol(major), density.