Ejercicio2_BD2

#variables Cualitativas nominal:  Track, Artista

#variables Cualitativas Discretas: Populary , year


#Variables Cuantitativas: Duration , 

##Graficado de base datos 1980s classic

library(dslabs)
library(readr)
library(dplyr)

## 
## Adjuntando el paquete: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

library(viridis)

## Cargando paquete requerido: viridisLite

library(viridisLite)
library(RColorBrewer)

library(ggplot2)

#base datos externa https://www.kaggle.com/datasets/thebumpkin/1980s-classic
bd_ana<-read.csv('1980sClassics.csv')

#Grafico de Torta
bd_ana <- bd_ana %>%
  mutate(Popularity_Category = cut(Popularity, breaks = 5, labels = paste0("Category ", 1:5)))

  tabla_Categoria <- table(bd_ana$Popularity_Category)
  porcentaje_categoria <- round(tabla_Categoria * 100 / sum(tabla_Categoria), 1)
  
  # Crear gráfico de torta con etiquetas de porcentaje
pie(porcentaje_categoria, 
    labels = paste(porcentaje_categoria, "%"), 
    main = "CATEGORIAS DE POPULARIDAD", 
    col = brewer.pal(4,"Accent"), 
    border = brewer.pal(4,"Accent"))

# Agregar leyenda
legend("bottomright", 
       legend = names(porcentaje_categoria), 
       fill = brewer.pal(4,"Accent"), 
       cex = 0.6, 
       border = brewer.pal(4,"Accent"))

#Grafico de Barras

artist_counts <- bd_ana %>%
  group_by(Artist) %>%
  summarise(Track_Count = n()) %>%
  arrange(desc(Track_Count)) %>%
  head(10)

# Crear el gráfico de barras
barplot(artist_counts$Track_Count,
        names.arg = artist_counts$Artist,
        main = "Gráfico de Barras Frecuencia por Artista",
        ylab = "Frecuencia",
        xlab = "Artista",
        ylim = c(0, 20),
        col = viridis(4),
        border=viridis(4),
        las = 2
        ) # rotar etiquetas del eje x para mejor lectura 

grid(nx = NA, ny = NULL, col = "lightgray", lty = "dotted", lwd = par("lwd"))

#Histograma

duracion_year <- bd_ana %>%
  mutate(Duration_Seconds = as.numeric(substr(Duration, 1, regexpr(":", Duration) - 1)) * 60 + 
                          as.numeric(substr(Duration, regexpr(":", Duration) + 1, nchar(Duration)))) %>%
  mutate(collab = grepl('1989', Year)) %>%
  filter(collab == TRUE)

hist(duracion_year$Duration_Seconds, 
     breaks = "Sturges", 
     main = "Histograma de Duracion Canciones año 1989", 
     xlab = "Duracion canciones", 
     col = brewer.pal(9,"Set1"), 
     border = brewer.pal(9,"Set1"), 
     ylim = c(0,40),
     xlim = c(0, 500),
     labels = TRUE)

# Histograma densidad
hist(duracion_year$Duration_Seconds, 
     breaks = "Sturges", 
     probability = TRUE, 
     main = "Histograma Densidad de Duracion Canciones año 1989", 
     xlab = "Duracion canciones", 
     col = brewer.pal(9,"Set1"), 
     border = brewer.pal(9,"Set1"), 
     ylim = c(0,0.008),
     xlim = c(0, 500))

lines(density(duracion_year$Duration_Seconds), col = "black", lwd = 2)

#Gráfico de Caja (Boxplot)

summary(duracion_year$Duration_Seconds)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    79.0   210.5   250.5   255.1   293.2   405.0

# Obtener los valores resumen del summary()
resumen <- summary(duracion_year$Duration_Seconds)

# Extraer los valores importantes del summary
minimo <- resumen["Min."]
primer_cuartil <- resumen["1st Qu."]
mediana <- resumen["Median"]
media <- mean(duracion_year$Duration_Seconds, na.rm = TRUE)
tercer_cuartil <- resumen["3rd Qu."]
maximo <- resumen["Max."]

# Graficar el boxplot
boxplot(duracion_year$Duration_Seconds, 
        col="Blue", 
        ylab="Tasa de duración de canciones", 
        outline=FALSE, 
        main="Boxplot", 
        ylim=c(0,450))

# Adicionar la media
points(media, col="black", pch=10)
text(media, paste("Media:", round(media, 2)), pos=4, cex=0.8)

# Valores de resumen
valores <- c(minimo, primer_cuartil, mediana, media, tercer_cuartil, maximo)
labels <- c("Min", "1st Qu.", "Mediana", "Media", "3rd Qu.", "Max.")

# Adicionar los puntos y etiquetas para cada valor del resumen
for (i in 1:length(valores)) {
  points(valores[i], col="red", pch=10)
  text(valores[i], labels[i], pos=4, cex=0.8)
}

tabla_cate <- table(bd_ana$Popularity_Category)

#Comparacion #Diagrama de dispersion

x<- duracion_year$Popularity 
y<-duracion_year$Duration_Seconds

plot(x,y,main="Popularidad vs Duracion canciones",col=duracion_year$Popularity_Category,pch=20,xlab = "Popularidad",ylab="Duracion Segundos")
legend("bottomright",legend=levels(duracion_year$Popularity_Category),fill = unique(duracion_year$Popularity_Category),cex=0.75)

#Boxplot Duration_Seconds vs Popularity_Category
boxplot(Duration_Seconds~Popularity_Category,data=duracion_year,col=brewer.pal(9,"Set1"),outline=FALSE,ylim=c(0,300),main="Boxplot Duraccion  vs Popularity_Category")
abline(h=mean(duracion_year$Duration_Seconds),col="blue",lwd=2)
text(paste("",round(mean(duracion_year$Duration_Seconds),2)),x=1,y=100,col="blue")

#unir reportes

par(mfrow=c(1,2))

x<- duracion_year$Popularity 
y<-duracion_year$Duration_Seconds

plot(x,y,main="Popularidad vs Duracion canciones",col=duracion_year$Popularity_Category,pch=20,xlab = "Popularidad",ylab="Duracion Segundos")
legend("bottomright",legend=levels(duracion_year$Popularity_Category),fill = unique(duracion_year$Popularity_Category),cex=0.75)

boxplot(Duration_Seconds~Popularity_Category,data=duracion_year,col=brewer.pal(9,"Set1"),outline=FALSE,ylim=c(0,300),main="Boxplot Duraccion  vs Popularity_Category")
abline(h=mean(duracion_year$Duration_Seconds),col="blue",lwd=2)
text(paste("",round(mean(duracion_year$Duration_Seconds),2)),x=1,y=100,col="blue")