Caricamento pacchetti

library(dplyr)
library(ggplot2)
library(knitr)

Parte 1 : Dati

I seguenti dataset sono stati prelevati dal sito WineEnthusiast https://www.winemag.com/?s=&drink_type=wine durante la settimana a partire dal 15 giugno 2017 e poi di nuovo il 22 novembre 2017 tramite la tecnica dello scraping o raschiamento. I 2 dataset scaricabili da : https://www.kaggle.com/datasets/zynicide/wine-reviews?datasetId=1442&sortBy=voteCount&select=winemag-data_first150k.csv , uniti e con l’eliminazione delle osservazioni duplicate contengono 170.520 record . La variabili che si prendono in considerazione nell’analisi sono :

country: Stato in cui si è prodotto il vino
points: Punteggio attributo al vino dai sommelier
price: Prezzo del vino
province: Territorio dello Stato
variety: Varietà del vino

Caricamento dati

winemag.data.130k.v2 <- read.csv("winemag-data-130k-v2.csv")
winemag.data_first150k <- read.csv("winemag-data_first150k.csv")

df1 <- winemag.data_first150k %>%
  select(country,description,points,price,province,region_1,region_2, variety, winery)

df2 <- winemag.data.130k.v2 %>%
  select(country,description,points,price,province,region_1,region_2, variety, winery)

df1 <- rbind(df1,df2)

Parte 2 : Esplorazione dei dati

Si eliminano le osservazioni duplicate e si nota che la variabile price ha 12838 valori mancanti:

df1 <- unique.data.frame(df1)
colSums(is.na(df1))

##     country description      points       price    province    region_1 
##           0           0           0       12838           0           0 
##    region_2     variety      winery 
##           0           0           0

La variabile points ha un andamento quasi “normale” o simile alla curva di Gauss per cui la mediana è pari quasi alla media, mentre la variabile price è distorta sulla sinistra per cui il 50% dei vini ha un prezzo inferiore alla mediana pari a 25 dollari . La correlazione tra le 2 variabili è pari al 42,63% infatti ci si aspetta che il prezzo del vino sia correlato al suo punteggio:

df1 %>%
  ggplot(aes(points)) +
  geom_histogram(bins = 50, fill="red")

options(warn=-1)
df1 %>%
  ggplot(aes(price)) +
  geom_histogram(bins = 50, fill="orange")

df3 <- df1[-which(is.na(df1$price)),]
cor(df3$points,df3$price)

## [1] 0.4268176

summary(df1$points)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   80.00   86.00   88.00   88.24   90.00  100.00

summary(df1$price)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##    4.00   16.00   25.00   34.66   40.00 3300.00   12838

VINI DEL MONDO PARAGONATI ALL’ITALIA

Gli stati con il più alto numero di recenzioni sono : USA, Italia,Francia, Spagna, Portogallo ecc. :

df1 %>%
    group_by(country) %>%
    summarise(Totale=n()) %>%
    mutate(country=reorder(country,Totale)) %>%
    ggplot(aes(country,Totale, fill=country)) +
    geom_bar(stat = "identity")+
    coord_flip() +
    geom_text(aes(label=Totale), hjust=0, size=2)+
    guides(fill="none")+
    ggtitle("Recensioni totali dei vini per Stato")

Gli stati con la media o mediana dei punteggi dei vini più alti sono : Regno Unito, Austria , Germania, Francia, Marocco, Italia

df1 %>%
  mutate(country = reorder(country,points)) %>%
  ggplot(aes(points,country, fill=country))+
  geom_boxplot()+
  guides(fill="none")

df1 %>%
    group_by(country) %>%
    summarise(media_point= mean(points)) %>%
    mutate(country=reorder(country,media_point)) %>%
    ggplot(aes(country,media_point, fill=country)) +
    geom_bar(stat = "identity")+
    coord_flip() +
    geom_text(aes(label=round(media_point,2)), hjust=0, size=2)+
    guides(fill="none")+
    ggtitle("")

VINI ITALIANI

Le regioni italiane con il più alto numero di recenzioni sono Toscana, Piemonte e Veneto

df1 <- df1 %>%
  filter(country=="Italy")

df1$province<- as.factor(df1$province)
df1$region_1<- as.factor(df1$region_1)
df1$region_2<- as.factor(df1$region_2)
df1$variety<- as.factor(df1$variety)
df1$winery<- as.factor(df1$winery)

df1 %>%
    group_by(province) %>%
    summarise(Totale=n()) %>%
    mutate(province=reorder(province,Totale)) %>%
    ggplot(aes(province,Totale, fill=province)) +
    geom_bar(stat = "identity")+
    coord_flip() +
    geom_text(aes(label=Totale), hjust=0, size=2)+
    ggtitle("Recensioni totali dei vini per Territorio")

Le regioni italiana con la media o mediana dei punteggi maggiore sono Piemonte, Toscana e Lombardia:

df1 %>%
  mutate(province = reorder(province,points)) %>%
  ggplot(aes(points,province, fill=province))+
  geom_boxplot()

df1 %>%
  filter(points>=92) %>%
  group_by(province, variety) %>%
  summarise(media_punti=mean(points) , .groups="keep") %>%
  #mutate(variety = reorder(variety,media_punti)) %>%
  ggplot(aes(variety,media_punti, fill=province))+
  geom_bar(stat = "identity")+
  coord_flip() +
  ggtitle("Varietà di vini per territorio",subtitle =  " con un punteggio in media superiore a 92")

df1 %>%
  ggplot(aes(points,price,colour=province))+
  geom_point()

df3<-df1 %>%
  filter(points>quantile(points, p=0.75), price<quantile(price, p=0.25, na.rm=TRUE)) %>%
  select(variety,points,price,province) %>%
  arrange(desc(points))

kable(df3, caption = "Vini italiani migliori a prezzo più basso")

Vini italiani migliori a prezzo più basso
variety	points	price	province
Red Blend	93	16	Tuscany
Sangiovese	92	13	Tuscany
Sauvignon	92	17	Northeastern Italy
Sangiovese	92	16	Tuscany
Verduzzo	92	14	Veneto
Red Blend	92	15	Veneto
Vermentino	92	17	Sicily & Sardinia
Carignano	92	15	Sicily & Sardinia
Moscato	92	16	Piedmont
Moscato	92	17	Piedmont
Red Blend	92	16	Piedmont
Turbiana	91	17	Lombardy
Malvasia	91	16	Northeastern Italy
Red Blend	91	15	Tuscany
Sangiovese	91	15	Tuscany
Sangiovese	91	16	Tuscany
Sangiovese Grosso	91	15	Tuscany
Red Blend	91	16	Tuscany
Red Blend	91	12	Tuscany
Red Blend	91	17	Sicily & Sardinia
Sangiovese	91	12	Tuscany
Merlot	91	17	Northeastern Italy
Red Blend	91	17	Southern Italy
Lambrusco di Sorbara	91	16	Central Italy
Glera	91	15	Northeastern Italy
Pinot Grigio	91	13	Northeastern Italy
Verdicchio	91	15	Central Italy
Sangiovese	91	12	Tuscany
Albana	91	15	Central Italy
White Blend	91	17	Veneto
Glera	91	13	Veneto
Pinot Bianco	91	13	Northeastern Italy
Glera	91	15	Northeastern Italy
Verdicchio	91	17	Central Italy
Red Blend	91	16	Tuscany
Turbiana	91	17	Lombardy
Verdicchio	91	17	Central Italy
Pinot Bianco	91	16	Northeastern Italy
Sauvignon Blanc	91	15	Northeastern Italy

df2<-df1 %>%
  filter(points>=92) %>%
  arrange(desc(points),province,variety) %>%
  select(variety,province,points,price)

df2<- df2 %>%
  group_by(variety,province,points) %>%
  summarise(median_price=median(price))

## `summarise()` has grouped output by 'variety', 'province'. You can override
## using the `.groups` argument.

df2 <- unique.data.frame(df2)


df2<-df2 %>%
  arrange(desc(points),province,variety)

kable(df2, caption = "Vini con il punteggio maggiore da 92 fino a 100")

Vini con il punteggio maggiore da 92 fino a 100
variety	province	points	median_price
Merlot	Tuscany	100	460.0
Prugnolo Gentile	Tuscany	100	210.0
Red Blend	Tuscany	100	195.0
Sangiovese	Tuscany	100	550.0
Sangiovese Grosso	Tuscany	100	270.0
Nebbiolo	Piedmont	99	440.0
Merlot	Tuscany	99	285.0
Prugnolo Gentile	Tuscany	99	237.0
Red Blend	Tuscany	99	217.5
Sangiovese	Tuscany	99	200.0
Nebbiolo	Piedmont	98	117.0
Cabernet Franc	Tuscany	98	127.5
Merlot	Tuscany	98	227.5
Red Blend	Tuscany	98	NA
Sangiovese	Tuscany	98	120.0
Sangiovese Grosso	Tuscany	98	245.0
Syrah	Tuscany	98	225.0
Picolit	Northeastern Italy	97	90.0
Nebbiolo	Piedmont	97	NA
Nerello Mascalese	Sicily & Sardinia	97	60.0
Aglianico	Southern Italy	97	90.0
Cabernet Franc	Tuscany	97	130.0
Merlot	Tuscany	97	277.5
Red Blend	Tuscany	97	217.5
Sangiovese	Tuscany	97	130.0
Sangiovese Grosso	Tuscany	97	79.5
Syrah	Tuscany	97	160.0
White Blend	Tuscany	97	NA
Sagrantino	Central Italy	96	80.0
Nebbiolo	Lombardy	96	105.0
Picolit	Northeastern Italy	96	90.0
Nebbiolo	Piedmont	96	NA
Merlot	Southern Italy	96	105.0
Cabernet Franc	Tuscany	96	105.0
Merlot	Tuscany	96	315.0
Red Blend	Tuscany	96	112.5
Sangiovese	Tuscany	96	NA
Sangiovese Grosso	Tuscany	96	75.0
White Blend	Tuscany	96	62.5
Corvina, Rondinella, Molinara	Veneto	96	NA
Montepulciano	Central Italy	95	95.0
Sagrantino	Central Italy	95	70.0
Verdicchio	Central Italy	95	57.0
White Blend	Central Italy	95	NA
Nebbiolo	Lombardy	95	82.0
Friulano	Northeastern Italy	95	80.0
Kerner	Northeastern Italy	95	25.0
Picolit	Northeastern Italy	95	100.0
Pinot Bianco	Northeastern Italy	95	165.0
Sauvignon	Northeastern Italy	95	40.0
Teroldego	Northeastern Italy	95	50.0
White Blend	Northeastern Italy	95	88.0
Nebbiolo	Piedmont	95	NA
Red Blend	Piedmont	95	350.0
Carricante	Sicily & Sardinia	95	45.0
Nerello Mascalese	Sicily & Sardinia	95	60.0
Red Blend	Sicily & Sardinia	95	30.0
Zibibbo	Sicily & Sardinia	95	40.0
Aglianico	Southern Italy	95	NA
Cabernet Franc	Tuscany	95	100.0
Cabernet Sauvignon	Tuscany	95	75.0
Cabernet Sauvignon-Merlot	Tuscany	95	50.0
Merlot	Tuscany	95	130.0
Red Blend	Tuscany	95	NA
Sangiovese	Tuscany	95	NA
Sangiovese Grosso	Tuscany	95	NA
Syrah	Tuscany	95	65.0
White Blend	Tuscany	95	NA
Corvina, Rondinella, Molinara	Veneto	95	127.5
Garganega	Veneto	95	31.0
Albana	Central Italy	94	NA
Montepulciano	Central Italy	94	100.0
Sagrantino	Central Italy	94	55.0
Verdicchio	Central Italy	94	56.0
White Blend	Central Italy	94	35.0
Nebbiolo	Italy Other	94	NA
Pinot Nero	Lombardy	94	NA
Sparkling Blend	Lombardy	94	75.0
Turbiana	Lombardy	94	24.0
Cabernet Sauvignon	Northeastern Italy	94	92.0
Chardonnay	Northeastern Italy	94	125.0
Friulano	Northeastern Italy	94	35.0
Gewürztraminer	Northeastern Italy	94	50.0
Nosiola	Northeastern Italy	94	47.0
Picolit	Northeastern Italy	94	140.0
Ribolla Gialla	Northeastern Italy	94	NA
Sauvignon	Northeastern Italy	94	40.0
Sparkling Blend	Northeastern Italy	94	67.0
Teroldego	Northeastern Italy	94	50.0
Verduzzo	Northeastern Italy	94	NA
White Blend	Northeastern Italy	94	88.0
Barbera	Piedmont	94	NA
Cabernet Sauvignon	Piedmont	94	224.0
Nebbiolo	Piedmont	94	NA
Pinot Nero	Piedmont	94	45.0
Red Blend	Piedmont	94	NA
Refosco	Piedmont	94	200.0
Sparkling Blend	Piedmont	94	34.0
Carignano	Sicily & Sardinia	94	91.0
Nerello Mascalese	Sicily & Sardinia	94	67.5
Nero d’Avola	Sicily & Sardinia	94	51.0
Red Blend	Sicily & Sardinia	94	49.0
Vermentino	Sicily & Sardinia	94	113.0
White Blend	Sicily & Sardinia	94	65.5
Zibibbo	Sicily & Sardinia	94	40.0
Aglianico	Southern Italy	94	NA
Falanghina	Southern Italy	94	NA
Fiano	Southern Italy	94	35.0
Greco	Southern Italy	94	60.0
Moscato	Southern Italy	94	29.0
Red Blend	Southern Italy	94	NA
Sirica	Southern Italy	94	50.0
Bordeaux-style Red Blend	Tuscany	94	145.0
Cabernet Franc	Tuscany	94	77.5
Cabernet Sauvignon	Tuscany	94	159.5
Merlot	Tuscany	94	250.0
Moscadello	Tuscany	94	55.0
Red Blend	Tuscany	94	NA
Sangiovese	Tuscany	94	NA
Sangiovese Grosso	Tuscany	94	NA
Syrah	Tuscany	94	96.0
Vernaccia	Tuscany	94	NA
White Blend	Tuscany	94	NA
Cabernet Sauvignon	Veneto	94	80.0
Corvina	Veneto	94	87.0
Corvina, Rondinella, Molinara	Veneto	94	NA
Garganega	Veneto	94	48.0
Glera	Veneto	94	NA
Red Blend	Veneto	94	85.0
White Blend	Veneto	94	31.0
Albana	Central Italy	93	27.5
Cabernet Blend	Central Italy	93	50.0
Cabernet Sauvignon	Central Italy	93	52.0
Chardonnay	Central Italy	93	60.0
Lambrusco di Sorbara	Central Italy	93	20.0
Moscato	Central Italy	93	45.0
Red Blend	Central Italy	93	49.0
Sagrantino	Central Italy	93	50.0
Trebbiano	Central Italy	93	50.0
Verdicchio	Central Italy	93	57.0
White Blend	Central Italy	93	NA
Glera	Italy Other	93	49.0
Nerello Mascalese	Italy Other	93	144.5
Chardonnay	Lombardy	93	79.0
Nebbiolo	Lombardy	93	60.0
Pinot Nero	Lombardy	93	NA
Sparkling Blend	Lombardy	93	NA
Turbiana	Lombardy	93	24.0
Cabernet Sauvignon	Northeastern Italy	93	65.0
Chardonnay	Northeastern Italy	93	90.0
Friulano	Northeastern Italy	93	60.0
Kerner	Northeastern Italy	93	30.0
Merlot	Northeastern Italy	93	NA
Moscato Giallo	Northeastern Italy	93	NA
Müller-Thurgau	Northeastern Italy	93	40.0
Picolit	Northeastern Italy	93	68.0
Pinot Nero	Northeastern Italy	93	50.0
Red Blend	Northeastern Italy	93	57.5
Ribolla Gialla	Northeastern Italy	93	117.5
Sauvignon	Northeastern Italy	93	NA
Sylvaner	Northeastern Italy	93	25.0
Teroldego	Northeastern Italy	93	48.0
Tocai	Northeastern Italy	93	35.0
White Blend	Northeastern Italy	93	NA
Barbera	Piedmont	93	31.0
Moscato	Piedmont	93	NA
Nebbiolo	Piedmont	93	NA
Pinot Nero	Piedmont	93	36.0
Red Blend	Piedmont	93	NA
Sparkling Blend	Piedmont	93	NA
Cabernet Sauvignon	Sicily & Sardinia	93	65.0
Carricante	Sicily & Sardinia	93	30.0
Grillo	Sicily & Sardinia	93	30.0
Malvasia	Sicily & Sardinia	93	40.0
Nerello Mascalese	Sicily & Sardinia	93	50.5
Nero d’Avola	Sicily & Sardinia	93	80.0
Red Blend	Sicily & Sardinia	93	NA
Syrah	Sicily & Sardinia	93	57.5
Vermentino	Sicily & Sardinia	93	NA
White Blend	Sicily & Sardinia	93	NA
Zibibbo	Sicily & Sardinia	93	42.0
Aglianico	Southern Italy	93	NA
Fiano	Southern Italy	93	35.0
Greco	Southern Italy	93	60.0
Merlot	Southern Italy	93	112.5
Moscato	Southern Italy	93	NA
Red Blend	Southern Italy	93	80.0
Roviello	Southern Italy	93	28.0
Sirica	Southern Italy	93	23.0
Susumaniello	Southern Italy	93	NA
White Blend	Southern Italy	93	19.0
Cabernet Franc	Tuscany	93	90.0
Cabernet Sauvignon	Tuscany	93	NA
Cabernet Sauvignon-Merlot	Tuscany	93	69.0
Malvasia	Tuscany	93	50.0
Merlot	Tuscany	93	75.0
Moscadello	Tuscany	93	NA
Petit Verdot	Tuscany	93	39.0
Red Blend	Tuscany	93	NA
Sangiovese	Tuscany	93	NA
Sangiovese Grosso	Tuscany	93	NA
Syrah	Tuscany	93	NA
Vernaccia	Tuscany	93	23.0
White Blend	Tuscany	93	NA
Corvina	Veneto	93	92.5
Corvina, Rondinella, Molinara	Veneto	93	NA
Garganega	Veneto	93	NA
Glera	Veneto	93	29.0
Red Blend	Veneto	93	80.0
White Blend	Veneto	93	31.0
Albana	Central Italy	92	47.5
Aleatico	Central Italy	92	50.0
Cabernet Sauvignon	Central Italy	92	60.0
Chardonnay	Central Italy	92	55.0
Lambrusco di Sorbara	Central Italy	92	24.0
Moscato	Central Italy	92	42.0
Passerina	Central Italy	92	NA
Red Blend	Central Italy	92	57.0
Sagrantino	Central Italy	92	NA
Sangiovese	Central Italy	92	40.0
Trebbiano	Central Italy	92	50.0
Verdicchio	Central Italy	92	50.0
White Blend	Central Italy	92	43.5
Aglianico	Italy Other	92	43.0
Centesimino	Italy Other	92	36.0
Falanghina	Italy Other	92	39.0
Nebbiolo	Italy Other	92	NA
Nerello Mascalese	Italy Other	92	26.0
Red Blend	Italy Other	92	34.0
Sangiovese	Italy Other	92	50.0
Sparkling Blend	Italy Other	92	NA
White Blend	Italy Other	92	40.0
Chardonnay	Lombardy	92	NA
Nebbiolo	Lombardy	92	NA
Pinot Nero	Lombardy	92	NA
Sparkling Blend	Lombardy	92	NA
Turbiana	Lombardy	92	22.5
Chardonnay	Northeastern Italy	92	38.0
Friulano	Northeastern Italy	92	23.0
Gewürztraminer	Northeastern Italy	92	39.0
Kerner	Northeastern Italy	92	25.0
Müller-Thurgau	Northeastern Italy	92	40.0
Picolit	Northeastern Italy	92	77.5
Pignolo	Northeastern Italy	92	70.0
Pinot Bianco	Northeastern Italy	92	35.0
Pinot Grigio	Northeastern Italy	92	35.0
Pinot Nero	Northeastern Italy	92	44.0
Pinot Noir	Northeastern Italy	92	50.0
Red Blend	Northeastern Italy	92	61.0
Riesling	Northeastern Italy	92	29.0
Sauvignon	Northeastern Italy	92	33.0
Sauvignon Blanc	Northeastern Italy	92	NA
Sparkling Blend	Northeastern Italy	92	60.0
Teroldego	Northeastern Italy	92	53.5
Veltliner	Northeastern Italy	92	40.0
Verduzzo	Northeastern Italy	92	18.0
Verduzzo Friulano	Northeastern Italy	92	49.0
White Blend	Northeastern Italy	92	NA
Barbera	Piedmont	92	NA
Dolcetto	Piedmont	92	NA
Moscato	Piedmont	92	NA
Nebbiolo	Piedmont	92	NA
Pinot Nero	Piedmont	92	52.0
Red Blend	Piedmont	92	38.0
Timorasso	Piedmont	92	60.0
Cabernet Sauvignon	Sicily & Sardinia	92	67.5
Cannonau	Sicily & Sardinia	92	27.0
Carignano	Sicily & Sardinia	92	NA
Carricante	Sicily & Sardinia	92	33.5
Chardonnay	Sicily & Sardinia	92	36.5
Fiano	Sicily & Sardinia	92	43.0
Grecanico	Sicily & Sardinia	92	44.0
Moscato	Sicily & Sardinia	92	40.0
Moscato di Noto	Sicily & Sardinia	92	40.0
Nasco	Sicily & Sardinia	92	NA
Nerello Cappuccio	Sicily & Sardinia	92	37.5
Nerello Mascalese	Sicily & Sardinia	92	41.0
Nero d’Avola	Sicily & Sardinia	92	42.0
Perricone	Sicily & Sardinia	92	26.0
Red Blend	Sicily & Sardinia	92	NA
Syrah	Sicily & Sardinia	92	48.5
Vermentino	Sicily & Sardinia	92	20.0
White Blend	Sicily & Sardinia	92	NA
Zibibbo	Sicily & Sardinia	92	NA
Aglianico	Southern Italy	92	NA
Caprettone	Southern Italy	92	19.0
Falanghina	Southern Italy	92	22.0
Fiano	Southern Italy	92	37.0
Greco	Southern Italy	92	38.0
Nero di Troia	Southern Italy	92	65.0
Primitivo	Southern Italy	92	NA
Red Blend	Southern Italy	92	NA
Roviello	Southern Italy	92	NA
Sirica	Southern Italy	92	23.0
Susumaniello	Southern Italy	92	50.0
Uva di Troia	Southern Italy	92	30.0
White Blend	Southern Italy	92	NA
Cabernet Blend	Tuscany	92	55.0
Cabernet Franc	Tuscany	92	110.0
Cabernet Sauvignon	Tuscany	92	NA
Cabernet Sauvignon-Merlot	Tuscany	92	61.5
Chardonnay	Tuscany	92	120.0
Merlot	Tuscany	92	NA
Moscadello	Tuscany	92	39.5
Petit Verdot	Tuscany	92	35.0
Pinot Nero	Tuscany	92	NA
Prugnolo Gentile	Tuscany	92	65.0
Red Blend	Tuscany	92	NA
Sangiovese	Tuscany	92	NA
Sangiovese Grosso	Tuscany	92	NA
Syrah	Tuscany	92	NA
Trebbiano-Malvasia	Tuscany	92	25.0
Vermentino	Tuscany	92	68.0
Vernaccia	Tuscany	92	22.0
White Blend	Tuscany	92	NA
Cabernet Sauvignon	Veneto	92	60.0
Chardonnay	Veneto	92	37.0
Corvina	Veneto	92	85.0
Corvina, Rondinella, Molinara	Veneto	92	NA
Durella	Veneto	92	40.0
Garganega	Veneto	92	NA
Glera	Veneto	92	NA
Red Blend	Veneto	92	NA
Sparkling Blend	Veneto	92	NA
Verduzzo	Veneto	92	14.0
Vespaiolo	Veneto	92	40.0

colore<-2

for (r in levels(df1$province)) {
  
g<-  df1 %>%
    filter(province==r) %>%
    group_by(variety) %>%
    summarise(media_point= mean(points)) %>%
    mutate(variety=reorder(variety,media_point)) %>%
    ggplot(aes(variety,media_point)) +
    geom_bar(stat = "identity", fill=colore)+
    coord_flip() +
    geom_text(aes(label=round(media_point,2)), hjust=0, size=2)+
    ggtitle(paste("Vini con punteggio più alto in",r))

  print(g)
  colore <- colore + 2
}

Statistiche sui vini italiani migliori

Gianni

2022-06-09