Práca s dátami

knitr::opts_chunk$set(
    echo = TRUE,
    message = FALSE,
    warning = FALSE
)

library(tidyverse)
library(kableExtra)
library(ggplot2)

# Načítanie datasetu
udaje <- read.csv("Travel dataset.csv", header = TRUE, sep = ",", dec = ".")

# Zobrazenie prvých riadkov a názvov stĺpcov
head(udaje)

colnames(udaje)

 [1] "Trip.ID"              "Destination"          "Start.date"          
 [4] "End.date"             "Duration..days."      "Traveler.name"       
 [7] "Traveler.age"         "Traveler.gender"      "Traveler.nationality"
[10] "Accommodation.type"   "Accommodation.cost"   "Transportation.type" 
[13] "Transportation.cost" 

# Počet riadkov a stĺpcov
nrow(udaje)

[1] 139

ncol(udaje)

[1] 13

# Základná štatistika
summary(udaje)

    Trip.ID      Destination         Start.date          End.date        
 Min.   :  1.0   Length:139         Length:139         Length:139        
 1st Qu.: 35.5   Class :character   Class :character   Class :character  
 Median : 70.0   Mode  :character   Mode  :character   Mode  :character  
 Mean   : 70.0                                                           
 3rd Qu.:104.5                                                           
 Max.   :139.0                                                           
                                                                         
 Duration..days.  Traveler.name       Traveler.age   Traveler.gender   
 Min.   : 5.000   Length:139         Min.   :20.00   Length:139        
 1st Qu.: 7.000   Class :character   1st Qu.:28.00   Class :character  
 Median : 7.000   Mode  :character   Median :31.00   Mode  :character  
 Mean   : 7.606                      Mean   :33.18                     
 3rd Qu.: 8.000                      3rd Qu.:38.00                     
 Max.   :14.000                      Max.   :60.00                     
 NA's   :2                           NA's   :2                         
 Traveler.nationality Accommodation.type Accommodation.cost Transportation.type
 Length:139           Length:139         Length:139         Length:139         
 Class :character     Class :character   Class :character   Class :character   
 Mode  :character     Mode  :character   Mode  :character   Mode  :character   
                                                                               
                                                                               
                                                                               
                                                                               
 Transportation.cost
 Length:139         
 Class :character   
 Mode  :character   
                    
                    
                    
                    

# Výber konkrétnych premenných (napr. krajina, trvanie cesty, náklady)
udaje %>%
  select(Destination, Traveler.name, Traveler.age) %>%
  head(10) %>%
  kable(caption = "Ukážka vybraných premenných z datasetu") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

Ukážka vybraných premenných z datasetu

Destination	Traveler.name	Traveler.age
London, UK	John Smith	35
Phuket, Thailand	Jane Doe	28
Bali, Indonesia	David Lee	45
New York, USA	Sarah Johnson	29
Tokyo, Japan	Kim Nguyen	26
Paris, France	Michael Brown	42
Sydney, Australia	Emily Davis	33
Rio de Janeiro, Brazil	Lucas Santos	25
Amsterdam, Netherlands	Laura Janssen	31
Dubai, United Arab Emirates	Mohammed Ali	39

# Vyber len cesty, ktoré trvali viac ako 5 dní a zorad ich podľa výdavkov
udaje %>%
  filter(Duration..days. > 5) %>%
  arrange(desc(Duration..days.)) %>%
  head(10) %>%
  kable(caption = "Najdrahšie cesty s trvaním nad 5 dní") %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

Najdrahšie cesty s trvaním nad 5 dní

Trip.ID	Destination	Start.date	End.date	Duration..days.	Traveler.name	Traveler.age	Traveler.gender	Traveler.nationality	Accommodation.type	Accommodation.cost	Transportation.type	Transportation.cost
4	New York, USA	8/15/2023	8/29/2023	14	Sarah Johnson	29	Female	British	Hotel	2000	Flight	1000
31	Australia	8/20/2022	9/2/2022	13	Emma Davis	28	Female	British	Hotel	$1,000	Car rental	$500
86	Bali	8/10/2021	8/20/2021	11	Maria Garcia	42	Female	Spanish	Resort	1200 USD	Plane	700 USD
89	London	11/20/2021	11/30/2021	11	James Wilson	29	Male	British	Hostel	300 USD	Plane	400 USD
92	Rome	3/10/2022	3/20/2022	11	Giulia Rossi	30	Female	Italian	Hostel	200 USD	Plane	350 USD
93	Bali	4/15/2022	4/25/2022	11	Putra Wijaya	33	Male	Indonesian	Villa	1500 USD	Car rental	300 USD
119	Sydney, Aus	5/1/2022	5/12/2022	11	Cindy Chen	26	Female	Chinese	Airbnb	800	Plane	1000
7	Sydney, Australia	11/20/2023	11/30/2023	10	Emily Davis	33	Female	Australian	Hostel	500	Flight	1200
18	Bali	8/15/2023	8/25/2023	10	Michael Chang	28	Male	Chinese	Resort	$1,500	Plane	$700
20	Tokyo	10/5/2023	10/15/2023	10	Kenji Nakamura	45	Male	Japanese	Hotel	$1,200	Plane	$800

ggplot(udaje, aes(x = Duration..days., y = Accommodation.cost, color = Transportation.cost)) +
  geom_point(alpha = 0.6) +
  theme_minimal() +
  labs(
    title = "Závislosť medzi dĺžkou cesty a výdavkami",
    x = "Dĺžka cesty (dni)",
    y = "Výdavky ($)"
  )

# Priemerné výdavky podľa typu dopravy
udaje %>%
  group_by(Transportation.type) %>%
  summarise(
    Priemerne_naklady = mean(Transportation.cost, na.rm = FALSE),
    Pocet_ciest = n()
  ) %>%
  kable(caption = "Priemerné výdavky podľa spôsobu dopravy") %>%
  kable_styling(bootstrap_options = c("striped", "hover"))

Priemerné výdavky podľa spôsobu dopravy

Transportation.type	Priemerne_naklady	Pocet_ciest
	NA	3
Airplane	NA	5
Bus	NA	6
Car	NA	3
Car rental	NA	13
Ferry	NA	1
Flight	NA	13
Plane	NA	57
Subway	NA	1
Train	NA	37

NA

udaje <- udaje %>%
  mutate(
    Cena_kategorie = case_when(
      Transportation.cost < 500 ~ "Nízke náklady",
      Transportation.cost < 1500 ~ "Stredné náklady",
      TRUE ~ "Vysoké náklady"
    )
  )

head(udaje)

# Scatterplot – závislosť medzi dĺžkou cesty a výdavkami
ggplot(udaje, aes(x = Duration..days., y = Accommodation.cost, color = Transportation.cost)) +
  geom_point(alpha = 0.6) +
  theme_minimal() +
  labs(
    title = "Závislosť medzi dĺžkou cesty a výdavkami",
    x = "Dĺžka cesty (dni)",
    y = "Výdavky ($)"
  )

This is an R Markdown Notebook. When you execute code within the notebook, the results appear beneath the code.

Try executing this chunk by clicking the Run button within the chunk or by placing your cursor inside it and pressing Ctrl+Shift+Enter.

plot(cars)

Add a new chunk by clicking the Insert Chunk button on the toolbar or by pressing Ctrl+Alt+I.

When you save the notebook, an HTML file containing the code and output will be saved alongside it (click the Preview button or press Ctrl+Shift+K to preview the HTML file).

The preview shows you a rendered HTML copy of the contents of the editor. Consequently, unlike Knit, Preview does not run any R code chunks. Instead, the output of the chunk when it was last run in the editor is displayed.