Chocolate Bar Rating
Chi Tran
12/29/2021
Context
Chocolate is one of the most popular candies in the world. This dataset contains expert ratings of over 1,700 individual chocolate bars, along with information on their regional origin, percentage of cocoa, the variety of chocolate bean used and where the beans were grown.
Flavors of Cacao Rating System:
5= Elite (Transcending beyond the ordinary limits)
4= Premium (Superior flavor development, character and style)
3= Satisfactory(3.0) to praiseworthy(3.75) (well made with special qualities)
2= Disappointing (Passable but contains at least one significant flaw)
1= Unpleasant (mostly unpalatable)
Data source
These ratings were compiled by Brady Brelinski, Founding Member of the Manhattan Chocolate Society.
The dataset is public at Kaggle
Findings
Which Companies produce the highest rating of chocolate?
Comparison of highest-rated Companies in term of Cocoa percentage.
What makes the best chocolate in term of rating and cocoa percentage?
Which countries produce the highest-rated bars?
Which countries are the largest producer of chocolate?
Where are the best cocoa beans grown?
What’s the relationship between cocoa solids percentage and rating?
Process
Installing and loading packages
library(tidyverse)## -- Attaching packages --------------------------------------- tidyverse 1.3.1 --## v ggplot2 3.3.5 v purrr 0.3.4
## v tibble 3.1.5 v dplyr 1.0.7
## v tidyr 1.1.4 v stringr 1.4.0
## v readr 2.0.2 v forcats 0.5.1## -- Conflicts ------------------------------------------ tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(janitor)## Warning: package 'janitor' was built under R version 4.1.2##
## Attaching package: 'janitor'## The following objects are masked from 'package:stats':
##
## chisq.test, fisher.testlibrary(ggplot2)
library(knitr)## Warning: package 'knitr' was built under R version 4.1.2Import Data
flavors_df <- read_csv("flavors_of_cacao.csv")## Rows: 1795 Columns: 9## -- Column specification --------------------------------------------------------
## Delimiter: ","
## chr (6): Company
## (Maker-if known), Specific Bean Origin
## or Bar Name, Cocoa
## ...
## dbl (3): REF, Review
## Date, Rating##
## i Use `spec()` to retrieve the full column specification for this data.
## i Specify the column types or set `show_col_types = FALSE` to quiet this message.Data summary
colnames(flavors_df)## [1] "Company \n(Maker-if known)" "Specific Bean Origin\nor Bar Name"
## [3] "REF" "Review\nDate"
## [5] "Cocoa\nPercent" "Company\nLocation"
## [7] "Rating" "Bean\nType"
## [9] "Broad Bean\nOrigin"glimpse(flavors_df)## Rows: 1,795
## Columns: 9
## $ `Company \n(Maker-if known)` <chr> "A. Morin", "A. Morin", "A. Morin"~
## $ `Specific Bean Origin\nor Bar Name` <chr> "Agua Grande", "Kpime", "Atsane", ~
## $ REF <dbl> 1876, 1676, 1676, 1680, 1704, 1315~
## $ `Review\nDate` <dbl> 2016, 2015, 2015, 2015, 2015, 2014~
## $ `Cocoa\nPercent` <chr> "63%", "70%", "70%", "70%", "70%",~
## $ `Company\nLocation` <chr> "France", "France", "France", "Fra~
## $ Rating <dbl> 3.75, 2.75, 3.00, 3.50, 3.50, 2.75~
## $ `Bean\nType` <chr> "Â ", "Â ", "Â ", "Â ", "Â ", "Criollo"~
## $ `Broad Bean\nOrigin` <chr> "Sao Tome", "Togo", "Togo", "Togo"~Data Cleaning
The data frame has unknown character in the column name so clean_names() function will be used to clean names.
cleaned_df <- clean_names(flavors_df)
glimpse(cleaned_df)## Rows: 1,795
## Columns: 9
## $ company_maker_if_known <chr> "A. Morin", "A. Morin", "A. Morin", "~
## $ specific_bean_origin_or_bar_name <chr> "Agua Grande", "Kpime", "Atsane", "Ak~
## $ ref <dbl> 1876, 1676, 1676, 1680, 1704, 1315, 1~
## $ review_date <dbl> 2016, 2015, 2015, 2015, 2015, 2014, 2~
## $ cocoa_percent <chr> "63%", "70%", "70%", "70%", "70%", "7~
## $ company_location <chr> "France", "France", "France", "France~
## $ rating <dbl> 3.75, 2.75, 3.00, 3.50, 3.50, 2.75, 3~
## $ bean_type <chr> "Â ", "Â ", "Â ", "Â ", "Â ", "Criollo", "~
## $ broad_bean_origin <chr> "Sao Tome", "Togo", "Togo", "Togo", "~colnames(cleaned_df)## [1] "company_maker_if_known" "specific_bean_origin_or_bar_name"
## [3] "ref" "review_date"
## [5] "cocoa_percent" "company_location"
## [7] "rating" "bean_type"
## [9] "broad_bean_origin"renamed_cleaned_df <- cleaned_df %>%
rename(Company = company_maker_if_known)glimpse(renamed_cleaned_df)## Rows: 1,795
## Columns: 9
## $ Company <chr> "A. Morin", "A. Morin", "A. Morin", "~
## $ specific_bean_origin_or_bar_name <chr> "Agua Grande", "Kpime", "Atsane", "Ak~
## $ ref <dbl> 1876, 1676, 1676, 1680, 1704, 1315, 1~
## $ review_date <dbl> 2016, 2015, 2015, 2015, 2015, 2014, 2~
## $ cocoa_percent <chr> "63%", "70%", "70%", "70%", "70%", "7~
## $ company_location <chr> "France", "France", "France", "France~
## $ rating <dbl> 3.75, 2.75, 3.00, 3.50, 3.50, 2.75, 3~
## $ bean_type <chr> "Â ", "Â ", "Â ", "Â ", "Â ", "Criollo", "~
## $ broad_bean_origin <chr> "Sao Tome", "Togo", "Togo", "Togo", "~trimmed_RC_df <- renamed_cleaned_df %>%
select(Company,cocoa_percent,rating,company_location)head(trimmed_RC_df)## # A tibble: 6 x 4
## Company cocoa_percent rating company_location
## <chr> <chr> <dbl> <chr>
## 1 A. Morin 63% 3.75 France
## 2 A. Morin 70% 2.75 France
## 3 A. Morin 70% 3 France
## 4 A. Morin 70% 3.5 France
## 5 A. Morin 70% 3.5 France
## 6 A. Morin 70% 2.75 Francesum(is.na(trimmed_RC_df$cocoa_percent))## [1] 0sum(is.na(trimmed_RC_df$rating))## [1] 0sum(is.na(trimmed_RC_df$company_location))## [1] 0trimmed_RC_df %>%
summarise(sd = sd(rating),max= max(rating))## # A tibble: 1 x 2
## sd max
## <dbl> <dbl>
## 1 0.478 5best_trimmed_RC_df <- trimmed_RC_df %>%
filter(cocoa_percent >= 80 & rating >= 3.75)head(best_trimmed_RC_df)## # A tibble: 6 x 4
## Company cocoa_percent rating company_location
## <chr> <chr> <dbl> <chr>
## 1 Chocolate Makers 80% 3.75 Amsterdam
## 2 Chocolate Tree, The 80% 3.75 Scotland
## 3 Ethereal 80% 3.75 U.S.A.
## 4 Potomac 82% 3.75 U.S.A.
## 5 Pralus 80% 4 France
## 6 Rogue 80% 3.75 U.S.A.Share
The companies produce the highest rating of chocolate
ggplot(data = best_trimmed_RC_df) +
geom_bar(mapping = aes(x = Company, fill = rating)) +
theme(axis.text.x = element_text(angle = 10)) +
labs(title = "Companies produce the highest rating of chocolate",x = "Company", y = "Count")
Figure 1 : Companies produce the highest rating of chocolate
Figure 1 shows that Companies Pralus and Soma produce the highest rating of chocolate.
Comparison by Companies
ggplot(data = best_trimmed_RC_df) +
geom_col(mapping = aes(x = Company, y = cocoa_percent)) +
facet_wrap(~Company)+
coord_flip()+
theme(axis.text.x = element_text(angle = 90)) +
labs(title = "Comparison by Companies", x ="Companies", y = "Cocoa Percent")
Figure 2 : Comparison in Cocoa Percent by Companies
Figure 2 shows the difference in Cocoa Percentage that Companies use in their chocolate. Company Videri and Company Soma use higher cocoa percentage compared to other companies, recorded at 90% and 88%, respectively.
What cocoa percent and rating determine the best chocolate
ggplot(data = best_trimmed_RC_df) +
geom_point(mapping = aes(x = cocoa_percent, y = rating)) +
theme(axis.text.x = element_text(angle = 90))+
labs(title = "Best Chocolates", x = "Cocoa Percent", y = "Rating")
Figure 3 : What cocoa percent and rating determine the best chocolate
Figure 3 shows that the best chocolate with the highest rating can be made with 80% cocoa percentage or 88% cocoa percentage.
The countries produce the highest rating of chocolate
ggplot(data = best_trimmed_RC_df) +
geom_bar(mapping = aes(x = company_location, alpha = rating))+
theme(axis.text.x = element_text(angle = 90)) +
labs(title = "Countries produce the highest rating of chocolate",x = "Countries", y = "Count")
Figure 4 : What countries produce the highest rating of chocolate
Figure 4 shows that France and Canada produce the highest rating of chocolate.
Which countries are the largest producer of chocolate
arranged_trimmed_RC_df <- trimmed_RC_df %>%
count(company_location) %>%
arrange(desc(n))
head(arranged_trimmed_RC_df)## # A tibble: 6 x 2
## company_location n
## <chr> <int>
## 1 U.S.A. 764
## 2 France 156
## 3 Canada 125
## 4 U.K. 96
## 5 Italy 63
## 6 Ecuador 54countries_most <- arranged_trimmed_RC_df %>%
filter(n > 60)head(countries_most)## # A tibble: 5 x 2
## company_location n
## <chr> <int>
## 1 U.S.A. 764
## 2 France 156
## 3 Canada 125
## 4 U.K. 96
## 5 Italy 63ggplot(data = countries_most) +
geom_col(mapping = aes(reorder(company_location,n),n))+
labs(title = "Which coutries is the largest producer of chocolate",x = "Countries", y = "Count")
Figure 5 : Which countries are the largest producer of chocolate
Figure 5 shows that USA is the largest producer of chocolate, followed by France and Canada.
The Best Bean Grow Origin
bean_df <- renamed_cleaned_df %>%
select(broad_bean_origin, bean_type, rating)head(bean_df)## # A tibble: 6 x 3
## broad_bean_origin bean_type rating
## <chr> <chr> <dbl>
## 1 Sao Tome  3.75
## 2 Togo  2.75
## 3 Togo  3
## 4 Togo  3.5
## 5 Peru  3.5
## 6 Venezuela Criollo 2.75best_bean_df <- bean_df %>%
filter(rating > 3.8)
head(best_bean_df)## # A tibble: 6 x 3
## broad_bean_origin bean_type rating
## <chr> <chr> <dbl>
## 1 Peru  4
## 2 Venezuela Trinitario 4
## 3 Peru  4
## 4 Papua New Guinea  4
## 5 Ecuador  4
## 6 Venezuela Criollo (Porcelana) 4ggplot(data = best_bean_df) + geom_bar(mapping = aes(x = broad_bean_origin, fill = bean_type)) +
labs(title = "The Best Bean Grow Origin") +
coord_flip()+
theme(axis.text.x = element_text(angle = 90,vjust = 0.7)) 
Figure 6 : The origin of Best Bean Grow
Figure 6 shows that bean grows best in Venezuela, Peru and Madagascar.
The relationship between Rating and Cocoa Percent
relation_cocoa_df <- renamed_cleaned_df %>%
select(rating, cocoa_percent)
head(relation_cocoa_df)## # A tibble: 6 x 2
## rating cocoa_percent
## <dbl> <chr>
## 1 3.75 63%
## 2 2.75 70%
## 3 3 70%
## 4 3.5 70%
## 5 3.5 70%
## 6 2.75 70%arranged_data_df <- relation_cocoa_df %>%
arrange(cocoa_percent)
head(arranged_data_df)## # A tibble: 6 x 2
## rating cocoa_percent
## <dbl> <chr>
## 1 1.75 100%
## 2 1.5 100%
## 3 1.5 100%
## 4 3.5 100%
## 5 1 100%
## 6 1.5 100%ggplot(data=arranged_data_df) +
geom_point(mapping = aes(x = rating, y = cocoa_percent)) +
labs(title = "The relationship between rating and cocoa percent", x = "Rating", y= "Cocoa Percent")
Figure 7 : The relationship between rating and cocoa percent
Figure 7 shows that no relationship between rating and cocoa percent.