#Passo 1 - carregar dados

load("~/Base_de_dados-master/Base_de_dados-master/df_pokemon.RData")
load("~/Base_de_dados-master/Base_de_dados-master/CARROS.RData")
library(corrplot)
## corrplot 0.92 loaded

#Passo 2 - transformar dados

CARROS$Tipodecombustivel <- ifelse(CARROS$Tipodecombustivel==0,"gas","alcool")
summary(CARROS$Tipodecombustivel)
##    Length     Class      Mode 
##        32 character character
CARROS$TipodeMarcha <- ifelse(CARROS$TipodeMarcha==0,"auto","manual")
summary(CARROS$TipodeMarcha)
##    Length     Class      Mode 
##        32 character character

#Passo 3 - visualizaçao de dados #Diagrama de dispersao para duas variaveis quantitativas

# Preço e Km/l
# linear negativa fraca
par(bg="lightyellow")
par(cex=1.0)
plot(CARROS$Kmporlitro,CARROS$Preco,pch=16, col="navy", main = "correlaçao entre km/l e preço",
     ylab = "preço do carro", xlab = "km/l")
abline(lsfit(CARROS$Kmporlitro,CARROS$Preco),col="red")

#Preço e HP
#quanto maior o hp, mais caro

plot(CARROS$HP,CARROS$Preco, pch=16, col="royalblue",
     main="Correlaçao entre preço e HP", ylab = "preço", xlab = "hp")
abline(lsfit(CARROS$HP,CARROS$Preco),col="red")

#Preço e peso
#mais pesado, mais caro

plot(CARROS$Peso,CARROS$Preco, pch=16, col="royalblue",
     main="Correlaçao entre preço e Peso", ylab = "preço", xlab = "Peso")
abline(lsfit(CARROS$Peso,CARROS$Preco),col="red")

#pokemon
#atk e speed mias rapido mais forte

par(bg="skyblue")
plot(df$attack,df$speed, pch= 16, col="red", main="atk e speed",
     ylab = "atk", xlab = "speed")
abline(lsfit(df$attack,df$speed),col="yellow")

#speed e def (zero correlaçao)

plot(df$defense,df$speed, pch= 16, col="red", main="def e speed",
     ylab = "def", xlab = "speed")
abline(lsfit(df$defense,df$speed),col="yellow")

#Passo 4 - Correlaçao entre variaveis

cor(df$defense,df$speed)
## [1] -0.00597676
cor(df$attack,df$speed)
## [1] 0.3356615
cor(CARROS$Peso,CARROS$Preco)
## [1] 0.8879799
cor(CARROS$Kmporlitro,CARROS$Preco)
## [1] -0.8475514
cor(CARROS$HP,CARROS$Preco)
## [1] 0.7909486

#Passo 5 - matriz de correlaçao

names(CARROS)
##  [1] "Kmporlitro"              "Cilindros"              
##  [3] "Preco"                   "HP"                     
##  [5] "Amperagem_circ_eletrico" "Peso"                   
##  [7] "RPM"                     "Tipodecombustivel"      
##  [9] "TipodeMarcha"            "NumdeMarchas"           
## [11] "NumdeValvulas"
names(df)
##  [1] "id"              "pokemon"         "species_id"      "height"         
##  [5] "weight"          "base_experience" "type_1"          "type_2"         
##  [9] "attack"          "defense"         "hp"              "special_attack" 
## [13] "special_defense" "speed"           "color_1"         "color_2"        
## [17] "color_f"         "egg_group_1"     "egg_group_2"     "url_image"      
## [21] "x"               "y"
Selecao = c("Kmporlitro","Cilindros","Preco","HP","Amperagem_circ_eletrico","Peso","RPM")

selecao2 <-c("height","weight","base_experience",  
             "attack","defense","hp","special_attack" ,
             "special_defense", "speed")


corcarros= cor(CARROS[,Selecao])
corcarros
##                         Kmporlitro  Cilindros      Preco         HP
## Kmporlitro               1.0000000 -0.8521620 -0.8475514 -0.7761684
## Cilindros               -0.8521620  1.0000000  0.9020329  0.8324475
## Preco                   -0.8475514  0.9020329  1.0000000  0.7909486
## HP                      -0.7761684  0.8324475  0.7909486  1.0000000
## Amperagem_circ_eletrico  0.6811719 -0.6999381 -0.7102139 -0.4487591
## Peso                    -0.8676594  0.7824958  0.8879799  0.6587479
## RPM                      0.4186840 -0.5912421 -0.4336979 -0.7082234
##                         Amperagem_circ_eletrico       Peso         RPM
## Kmporlitro                           0.68117191 -0.8676594  0.41868403
## Cilindros                           -0.69993811  0.7824958 -0.59124207
## Preco                               -0.71021393  0.8879799 -0.43369788
## HP                                  -0.44875912  0.6587479 -0.70822339
## Amperagem_circ_eletrico              1.00000000 -0.7124406  0.09120476
## Peso                                -0.71244065  1.0000000 -0.17471588
## RPM                                  0.09120476 -0.1747159  1.00000000
corpkmn= cor(df[,selecao2])
corpkmn
##                    height    weight base_experience    attack     defense
## height          1.0000000 0.6465813       0.4804062 0.4088367  0.35995909
## weight          0.6465813 1.0000000       0.4874248 0.4605428  0.48171259
## base_experience 0.4804062 0.4874248       1.0000000 0.5891011  0.50577735
## attack          0.4088367 0.4605428       0.5891011 1.0000000  0.43177454
## defense         0.3599591 0.4817126       0.5057773 0.4317745  1.00000000
## hp              0.4401011 0.4314012       0.6849912 0.4298658  0.23521065
## special_attack  0.3330252 0.2793074       0.6630866 0.3278213  0.19828306
## special_defense 0.3239820 0.3403940       0.6723205 0.2008107  0.48187370
## speed           0.2249439 0.1081207       0.5021202 0.3356615 -0.00597676
##                        hp special_attack special_defense       speed
## height          0.4401011      0.3330252       0.3239820  0.22494390
## weight          0.4314012      0.2793074       0.3403940  0.10812069
## base_experience 0.6849912      0.6630866       0.6723205  0.50212021
## attack          0.4298658      0.3278213       0.2008107  0.33566149
## defense         0.2352107      0.1982831       0.4818737 -0.00597676
## hp              1.0000000      0.3678422       0.3838715  0.16941766
## special_attack  0.3678422      1.0000000       0.4867300  0.44706699
## special_defense 0.3838715      0.4867300       1.0000000  0.23825607
## speed           0.1694177      0.4470670       0.2382561  1.00000000
par(bg="lightyellow")
par(cex=0.8)
corrplot(corcarros)

par(cex=0.6)
corrplot(corpkmn, method = "number")