Exercitii

Cu tabelul pseudo_facebook.tsv:

str(pf)

'data.frame':   99003 obs. of  15 variables:
 $ userid               : int  2094382 1192601 2083884 1203168 1733186 1524765 1136133 1680361 1365174 1712567 ...
 $ age                  : int  14 14 14 14 14 14 13 13 13 13 ...
 $ dob_day              : int  19 2 16 25 4 1 14 4 1 2 ...
 $ dob_year             : int  1999 1999 1999 1999 1999 1999 2000 2000 2000 2000 ...
 $ dob_month            : Factor w/ 12 levels "ian","feb","mar",..: 11 11 11 12 12 12 1 1 1 2 ...
 $ gender               : Factor w/ 2 levels "female","male": 2 1 2 1 2 2 2 1 2 2 ...
 $ tenure               : int  266 6 13 93 82 15 12 0 81 171 ...
 $ friend_count         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ friendships_initiated: int  0 0 0 0 0 0 0 0 0 0 ...
 $ likes                : int  0 0 0 0 0 0 0 0 0 0 ...
 $ likes_received       : int  0 0 0 0 0 0 0 0 0 0 ...
 $ mobile_likes         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ mobile_likes_received: int  0 0 0 0 0 0 0 0 0 0 ...
 $ www_likes            : int  0 0 0 0 0 0 0 0 0 0 ...
 $ www_likes_received   : int  0 0 0 0 0 0 0 0 0 0 ...

1. Convertiti variabila dob_month in variabila de tip factor, cu etichetele corespunzatoare numelor lunilor anului.

pf <- read.csv('pseudo_facebook.tsv',sep='\t')
str(pf)

'data.frame':   99003 obs. of  15 variables:
 $ userid               : int  2094382 1192601 2083884 1203168 1733186 1524765 1136133 1680361 1365174 1712567 ...
 $ age                  : int  14 14 14 14 14 14 13 13 13 13 ...
 $ dob_day              : int  19 2 16 25 4 1 14 4 1 2 ...
 $ dob_year             : int  1999 1999 1999 1999 1999 1999 2000 2000 2000 2000 ...
 $ dob_month            : int  11 11 11 12 12 12 1 1 1 2 ...
 $ gender               : Factor w/ 2 levels "female","male": 2 1 2 1 2 2 2 1 2 2 ...
 $ tenure               : int  266 6 13 93 82 15 12 0 81 171 ...
 $ friend_count         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ friendships_initiated: int  0 0 0 0 0 0 0 0 0 0 ...
 $ likes                : int  0 0 0 0 0 0 0 0 0 0 ...
 $ likes_received       : int  0 0 0 0 0 0 0 0 0 0 ...
 $ mobile_likes         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ mobile_likes_received: int  0 0 0 0 0 0 0 0 0 0 ...
 $ www_likes            : int  0 0 0 0 0 0 0 0 0 0 ...
 $ www_likes_received   : int  0 0 0 0 0 0 0 0 0 0 ...

pf <- pseudo_facebook

is.factor(pf$dob_month)

[1] TRUE

str(pf)

'data.frame':   99003 obs. of  15 variables:
 $ userid               : int  2094382 1192601 2083884 1203168 1733186 1524765 1136133 1680361 1365174 1712567 ...
 $ age                  : int  14 14 14 14 14 14 13 13 13 13 ...
 $ dob_day              : int  19 2 16 25 4 1 14 4 1 2 ...
 $ dob_year             : int  1999 1999 1999 1999 1999 1999 2000 2000 2000 2000 ...
 $ dob_month            : Factor w/ 12 levels "ian","feb","mar",..: 11 11 11 12 12 12 1 1 1 2 ...
 $ gender               : Factor w/ 2 levels "female","male": 2 1 2 1 2 2 2 1 2 2 ...
 $ tenure               : int  266 6 13 93 82 15 12 0 81 171 ...
 $ friend_count         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ friendships_initiated: int  0 0 0 0 0 0 0 0 0 0 ...
 $ likes                : int  0 0 0 0 0 0 0 0 0 0 ...
 $ likes_received       : int  0 0 0 0 0 0 0 0 0 0 ...
 $ mobile_likes         : int  0 0 0 0 0 0 0 0 0 0 ...
 $ mobile_likes_received: int  0 0 0 0 0 0 0 0 0 0 ...
 $ www_likes            : int  0 0 0 0 0 0 0 0 0 0 ...
 $ www_likes_received   : int  0 0 0 0 0 0 0 0 0 0 ...

2. Reprezentati histogramele variabilelor likes, www_likes si mobile_likes.

library(ggplot2)
library(gridExtra)

package <U+393C><U+3E31>gridExtra<U+393C><U+3E32> was built under R version 3.5.3

g1 <- qplot(pf$likes)
g2 <- qplot(pf$www_likes)
g3 <- qplot(pf$mobile_likes)

g1

g2

g3

3. Grupati histogramele de la punctul 2 intr-un singur grafic, asezate pe o linie.

grid.arrange(g1,g2,g3, ncol=3)

4. Separati histogramele de la punctul 2 in functie de luna nasterii utilizatorilor in 12 grafice.

g1 <- qplot(likes, data=pf)+
  facet_wrap(~dob_month, ncol=4)
g2 <- qplot(www_likes, data=pf)+
  facet_wrap(~dob_month, ncol=4)
g3 <- qplot(mobile_likes, data=pf)+
  facet_wrap(~dob_month, ncol=4)

g1

g2

g3

5. Grupati cele 3 grafice de la punctul 4 intr-un singur grafic, asezate pe o coloana unul sub altul.

options(repr.plot.width=5, repr.plot.height=20) 
grid.arrange(g1,g2,g3,ncol=1)

6. Reprezentati variabilele likes, www_likes si mobile_likes prin 3 grafice de tip boxplot.

g4 <- qplot(y=www_likes, data=pf, geom="boxplot")

g4

7. Reprezentati variabilele likes, www_likes si mobile_likes prin 3 grafice de tip boxplot cu boxplot-uri diferite pentru luni de nastere diferite.

qplot(y=www_likes, data=pf, x=dob_month, geom="boxplot")

8. Reglati axele graficelor de la punctele 6 si 7 pentru a se putea vizualiza corpul boxplot-urilor.

qplot(y=www_likes, data=pf, x=dob_month, geom="boxplot")+
  coord_cartesian(ylim=c(0,20))

9. Reprezentati variabilele likes, www_likes si mobile_likes prin 3 grafice de tip linii poligonale.

qplot(x=www_likes, data=pf, geom="freqpoly")

10. Separati liniile poligonale de la cele trei grafice de la punctul 9 in functie de luna nasterii folosind culori diferite.

qplot(x=www_likes, data=pf, geom="freqpoly", color=dob_month)