---
tags: tekstiaine
---

# Jagatud märkmed (tekstitöötlus 2022)

## 13.04.2021

Arvutiklassis installiks

install.packages('rlang')
install.packages('tidyverse')
install.packages('tidytext')

Control + Shift + M korraga teeb %>% toru.

gapminder %>% 
  filter(country=="France")
  
gapminder %>% 
  filter(!country=="Germany")

gapminder %>% 
  filter(country!="Germany")



brasiilia <- gapminder %>%
  filter(country=="Brazil")


ameerika <- gapminder %>% 
  filter(continent=="Americas")


okeaania <- gapminder %>% 
  filter(continent=="Oceania")

gapminder %>% 
  filter(country=="Finland") %>% 
  arrange(desc(year))

?filter

gapminder %>% 
  filter(continent=="Europe") %>% 
  filter(year==1952) %>% 
  arrange(country,by_name=T)

järjestus <- gapminder %>% 
  filter(continent=="Europe") %>% 
  filter(year==1952) %>% 
  arrange(country)

gapminder %>%
  filter(continent=="Europe") %>% 
  filter(year==1952) %>% 
  select(gdpPercap, lifeExp)
  
  
gapminder %>% 
  mutate(nr = 1:1704)

gapminder %>%
  filter(country=="Brazil") %>%
  ggplot(aes(x=year,y=gdpPercap))+
  geom_point()+
  labs(title="Brasiilia rikkus 1952-2007")









## 20.04.2022

edetabel %>% 
  filter(rank %in% 6:10) %>% 
  filter(artist=="Smilers") %>% 
  select(year,song,rank)



edetabel %>%
  filter(artist=="Smilers") %>% 
  select(year,song,rank) %>% 
  filter(rank %in% 6:10)

edetabel %>%
  filter(artist=="Smilers" & rank %in% 6:10) %>% 
  select(year,song,rank)


edetabel %>%
  filter(artist=="Smilers" & rank > 5 &  rank < 11) %>% 
  select(year,song,rank)

edetabel %>%
filter(artist=="Maarja") %>%
count(year)

edetabel %>% 
  filter(language=="en") %>% 
  filter(year<2000) %>% 
  select(year,rank,song)

edetabel %>% 
  filter(language=="en") %>% 
  filter(year%in% 1990:1999) %>% 
  select(year,rank,song)

edetabel %>% 
  filter(rank==1) %>% 
  sample_n(10) %>% 
  select(rank,year,song)

edetabel %>% 
  filter(artist=="Nancy") %>% 
  count(year)



edetabel %>% 
  group_by(artist) %>% 
  mutate(first=min(year), last=max(year)) %>% 
  filter(first>2000)

edetabel %>% 
  group_by(artist) %>% 
  arrange(artist, rank) %>% 
  mutate(artistbest=row_number()) %>% 
  mutate(lastplace=max(artistbest)) %>% 
  filter(artistbest==lastplace) %>% 
  ungroup()

## 27.04.2022

stopsonad <- read_tsv("https://datadoi.ee/bitstream/handle/33/78/estonian-stopwords.txt?sequence=1&isAllowed=y",col_names = "word")

laulusonad %>% 
  left_join(stopsonad2,by="word") %>% 
  group_by(artist) %>% 
  filter(artist=="Nublu") %>% 
  count(onstopsona,word,sort=T) %>% 
  mutate(proportsioon= n/sum(n)) %>% 
  filter(onstopsona==T) %>% 
  arrange(desc(proportsioon))

NA==NA
is.na(NA)
is.na("NA")

ilukirj_sonad <- read_tsv("http://datadoi.ee/bitstream/handle/33/41/token_1_grams.tsv?sequence=4&isAllowed=y",col_names = c("word","n_token","n_docs"))



ilukirj_sonad <- ilukirj_sonad %>%
  mutate(prop_ilukirj=n_token/sum(n_token))

laulusonad %>%
  filter(language=="et") %>%
  group_by(artist, song) %>%
  count(word,sort=T) %>%
  filter(!word %in% c("ref")) %>%
  mutate(proportsioon=n/sum(n)) %>%
  arrange(desc(n))

laulusonad %>%
  filter(language=="et") %>%
  anti_join(stopsonad,by="word") %>% 
  group_by(artist, song) %>%
  count(word,sort=T) %>%
  filter(!word %in% c("ref")) %>%
  mutate(proportsioon=n/sum(n)) %>%
  arrange(desc(n))

laulusonad %>% 
  group_by(artist,song) %>% 
  mutate(rownr=row_number()) %>% 
  mutate(n=n()) %>% 
  mutate(asukoht=rownr/n) %>% 
  filter(artist=="reket") %>% 
  filter(word=="ei")

## 04.05.2022

### Ühe artisti laulude ülevaade, graafikud:
Jam



Terminaator

### Mõne sõna asukohad mõnes loos


olla ja saada, võimalik, et liiga palju


### Regulaaravaldistest

"me oleme siin" %>% str_detect("me")
"me oleme siin" %>% str_detect("ma")
"me oleme siin" %>% str_detect("^me")
"me oleme siin" %>% str_detect(" me ")
"me oleme siin" %>% str_extract("me")
"me oleme siin" %>% str_extract("ma")
"me oleme siin" %>% str_extract_all("me")
"me oleme siin" %>% str_extract("me")



## 11.05.2022






top5grams <- viisgrammid %>% 
  count(fivegram,sort=T) %>% 
  mutate(rownr=row_number()) %>% 
  filter(rownr<11)



asukohad_viisgrammid <- viisgrammid %>% 
  mutate(nr=row_number(), n=n()) %>% 
  mutate(asukoht=nr/(n+1)) %>% 
  ungroup()

asukohad_viisgrammid %>% 
  inner_join(top5grams,by="fivegram") %>% 
  ggplot(aes(x=asukoht,y=fivegram))+
  geom_point()

tf_idfs_peatykid %>% 
  mutate(row_number=row_number()) %>% 
  filter(row_number<11) %>% 
  filter(chapter>9) %>% 
  filter(chapter<21) %>% 
  ggplot(aes(x=chapter,y=row_number,label=word))+
  geom_label()

words %>% 
  count(filename) %>% 
  ggplot(aes(y=filename,x=n))+
  geom_col()

words %>% 
  count(word,sort=T)



words %>% 
  count(word,sort=T) %>% 
  filter(row_number()<11) %>% 
  ggplot(aes(y=word,x=n))+
  geom_col()

words %>% 
  group_by(filename) %>% 
  count(word,sort=T)

words %>% 
  group_by(filename) %>% 
  count(word,sort=T) %>% 
  filter(row_number()<11) %>% 
  filter(filename=="Andres_Saal_Vambola.utf8")
  

words %>% 
  group_by(filename) %>% 
  count(word,sort=T) %>% 
  filter(row_number()<11)