#tidytuesday:Eurovision (2022 Final Televote Votes)

This is the clustering for the televotes:
See the group itself one for Ukraine, and see the case for Serbia.

#tidytuesday:Eurovision (2022 Final Jury Votes)

There is some polemic with the votes in Eurovision Final 2022 with some agreements possibly within some countries. Data Science can help us to see patterns within the countries, checking the clusters and with that base investigate more.

I show the cluster dendogram for the Final Votes from the Jury this year 2022.

#tidytuesday: Eurovision (All the winners from 1956 to 2022)

#tidytuesday:Eurovision (Spanish Artist and their positions)

 As we see in the plot Conchita Velasco and Raphael repeat representing Spain on Eurovision. The colour in the plots show their position (greener the better).

spanish_artists <-
eurovision_1 %>%
filter(section=="final"| section=="grand-final") %>%
filter(artist_country == "Spain") %>%
select(year, artist, song, host_city, host_country, rank)


spanish_artists %>%
group_by(year) %>%
mutate(artist = fct_reorder(artist, year)) %>%
ggplot(aes(x = year, y = artist)) +
geom_tile(aes(fill = rank), colour = "red") +
scale_fill_gradient(low="green", high="red", limits=c(1, 26)) +
scale_x_continuous(breaks = seq(1956, 2022)) +
theme(axis.text.x = element_text
                   (angle = 90, vjust = 0.5, hjust = 1)) +
labs(y = "Spanish artist on Eurovision by year", y = "Year")

#tidytuesday:Eurovision (What happened in 1956?)

Another special year. It was the first edition of Eurovision Festival with seven participant countries (founders) and two artists representing each country with different songs, except  Switzerland and Luxemburg with just one singer  for the two songs. 

Lys Assia from Switzerland was the winner.

There is a nice explanation on Wikipedia about that edition.

eurovision_1 %>%
    filter(year == 1956) %>%
    count(artist_country, sort = TRUE)
 A tibble: 7 × 2
artist_country               n
<chr>                      <int>
Belgium                      2
France                       2
Germany                      2
Italy                        2
Luxembourg                   2
Netherlands                  2
Switzerland                  2

eurovision_1 %>% 

+   filter(year == 1956) %>% 

+   count(artist, artist_country, song, winner, sort = TRUE) 

# A tibble: 14 × 5

   artist                 artist_country song                            winner     n

   <chr>                  <chr>          <chr>                           <lgl>  <int>

 1 Corry Brokken          Netherlands    Voorgoed Voorbij                FALSE      1

 2 Dany Dauberson         France         Il Est Là                       FALSE      1

 3 Franca Raimondi        Italy          Aprite Le Finestre              FALSE      1

 4 Freddy Quinn           Germany        So Geht Das Jede Nacht          FALSE      1

 5 Fud Leclerc            Belgium        Messieurs Les Noyés De La Seine FALSE      1

 6 Jetty Paerl            Netherlands    De Vogels Van Holland           FALSE      1

 7 Lys Assia              Switzerland    Das Alte Karussell              FALSE      1

 8 Lys Assia              Switzerland    Refrain                         TRUE       1

 9 Mathé Altéry           France         Le Temps Perdu                  FALSE      1

10 Michèle Arnaud         Luxembourg     Les Amants De Minuit            FALSE      1

11 Michèle Arnaud         Luxembourg     Ne Crois Pas                    FALSE      1

12 Mony Marc              Belgium        Le Plus Beau Jour De Ma Vie     FALSE      1

13 Tonina Torielli        Italy          Amami Se Vuoi                   FALSE      1

14 Walter Andreas Schwarz Germany        Im Wartesaal Zum Großen Glück   FALSE      1

All this must  be taken into account when working with some statistics with the Eurovision data base.

#tidytuesday:Eurovision (What happened in 1969?)

It was a special year with four winners:

eurovision_1 %>%
filter(section=="final"| section=="grand-final", winner == TRUE) %>%
count(winner, year, host_country, sort = TRUE)

# A tibble: 66 × 4

   winner  year host_country       n

   <lgl>  <dbl> <chr>          <int>

 1 TRUE    1969 Spain              4

 2 TRUE    1956 Switzerland        1

 3 TRUE    1957 Germany            1

 4 TRUE    1958 Netherlands        1

 5 TRUE    1959 France             1

 6 TRUE    1960 United Kingdom     1

 7 TRUE    1961 France             1

 8 TRUE    1962 Luxembourg         1

 9 TRUE    1963 United Kingdom     1

10 TRUE    1964 Denmark            1

# … with 56 more rows

 50 years ago today: Four winners at Eurovision 1969 in Madrid

#tidytuesday: Eurovision

 Another question we can ask about the history of Eurovision Festival is: which was the country which host the festival more often?

Normally the host country is the one which won the previous year, but for different reasons some years is not like that. That will be the case for 2023 where the host country will be United Kingdom (2nd place in 2022) indeed Ukraine for obvious reasons (1st place in 2022). 

As we can see in the plot UK is the country which host more ofen the Eurovision Festival.

countries_events <-
unique(eurovision_1 %>%
group_by(host_country) %>%
summarize(host_country, event, year))


countries_events %>%
count(host_country, sort = TRUE) %>%
ggplot(aes(x = reorder(host_country, n), n)) +
geom_col() +
coord_flip()

#tidytuesday: Eurovision

From today there is a new label in the blog to work with the tidytuesday data available from tidytuesday.

There are a lot of tutorials to learn to tidy our data to prepare it  to extract information, and to develop models (I normally see the videos of Julia Silge and David Robinson), so even if these post are not much related to the world of chemometrics they will help us to have fun with the "R" language, get better skills and habits that we will use it when necessary to work with R in our daily work.

I select the Eurovision data in this case, and the first thing I ask myself was to try to find a plot which shows me the evolution of the Eurovision contest

eurovision_1 %>%
    count(year,section, sort = TRUE) %>%
    ggplot(aes(year, n, fill = section )) +
    geom_col(width = 1, colour = "black") +
    theme(axis.text.x = element_text(angle = 90, vjust = 0.1)) +
    scale_x_continuous(breaks = seq(1956, 2022)) +
    theme(axis.text=element_text(size=7))  +
    coord_flip()

The plot shows the different periods of the Eurovision contest seeing how the number of countries increased almost every year. We see when the system of semi-finals started and several curious things, that we will try to discover in the label: tidytuesday:Eurovision.