---
tags: tekstiaine
---

# Tunni jooksul tehtud märkmed (SVUH.00.093 22.04.2023)

Koondan siia tunni jooksul tehtud märkmed kergemaks kopeerimiseks

Siia saab 
koodi kirjutada

Peetri näidis

Triin

Roosi

Hanna

Marleen

Elika

Ingeliis

Carmen

Vivian

asukohad %>% 
  filter(str_detect(word,"peremees|peremehe")) %>% 
  mutate(type=str_extract(word,"pereme")) %>% 
  ggplot(aes(x=asukoht,y=type))+
  geom_point()

asukohad %>% 
  filter(str_detect(word,"^so[oh]")) %>% 
  mutate(type=str_extract(word,"pereme")) %>% 
  ggplot(aes(x=asukoht,y=type))+
  geom_point()

### viisgrammid

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

viis_asukohad %>% 
  filter(fivegram=="vurr vurr vurr vurr vurr"|fivegram=="nõnda kestis see tükk aega") %>% 
  ggplot(aes(x=asukoht,y=fivegram))+
  geom_point()
  
  asukohad2 <- viisgrammid %>% 
  mutate(nr=row_number(), n=n()) %>% 
  mutate(asukoht=nr/(n+1)) %>% 
  ungroup() %>% 
  top_n(10)
  
  
  viisgrammid %>% 
  mutate(nr=row_number(), n=n()) %>% 
  mutate(asukoht=nr/(n+1)) %>% 
  filter(asukoht <= 10) %>% 
  ggplot(aes(x=asukoht,y=fivegram))+
  geom_point()


asukohad_valik <- viisgrammid %>% 
  mutate(nr=row_number(), n=n()) %>% 
  mutate(asukoht=nr/(n+1)) %>% 
  group_by(fivegram) %>% 
  mutate(word_n=n()) %>% 
  ungroup() %>% 
  filter(word_n>3)

asukohad_valik %>% 
  ggplot(aes(x=asukoht,y=fivegram))+
  geom_point()


peatykid_sonad2 <- raamat1 %>% 
  group_by(chapter) %>% 
  unnest_tokens(word,txt,to_lower=F) %>% 
  count(word,sort=T) %>% 
  mutate(sagedus=n()/sum(n))
  


peatykid_sonad2 %>% 
  anti_join(stopwords,"word") %>% 
  filter(!str_detect(word,"[A-ZÕÄÖÜ]")) %>%
  mutate(row_number=row_number()) %>% 
  filter(row_number<11) %>% 
  filter(chapter<11) %>% 
  ggplot(aes(x=chapter,y=row_number,label=word))+
  geom_label()
  
  
tf_idfs_peatykid %>% 
  mutate(row_number=row_number()) %>% 
  filter(row_number<11) %>% 
  filter(chapter>= 10 )  %>%
  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(filename) %>% 
  mutate(sõnad_raamatus=n)

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

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

words %>% 
  filter(filename=="Elisabeth_Aspe_Ennosaare_Ain.utf8") %>% 
  count(word, sort=T) %>% 
  mutate(nr=row_number()) %>% 
  filter(nr<11) %>% 
  ggplot(aes(y=word,x=n))+
  geom_col()

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

words %>% 
  group_by(filename) %>% 
  count(word,sort=T) %>% 
  mutate(freq=n/sum(n))%>% 
  filter(word == "tüdruk") %>%  
  ggplot(aes(y=filename,x=freq))+
  geom_point()+
  guides(color=F)

words %>% 
  group_by(filename) %>% 
  count(word,sort=T) %>% 
  mutate(freq=n/sum(n))%>% 
  filter(word=="tüdruk") %>% 
  ggplot(aes(y=filename,x=freq))+
  geom_point()+
  guides(color=F)

words %>% 
  group_by(filename) %>% 
  filter(str_detect(filename,"^Juri")) %>% 
  count(word,sort=T)  %>% 
  mutate(freq=n/sum(n)) %>% 
  ungroup() %>% 
  mutate(searchterm=str_extract(word,"^kirik|^kõrts")) %>% 
  #filter(searchterm=="kirik"|searchterm=="kõrts")+
  filter(!is.na(searchterm)) %>%  # kasutame ainult ridu, kus on olemas info searchterm tulbas
  group_by(filename,searchterm) %>% 
  summarise(freq=sum(freq)) %>% # tekstid on grupeeritud failinime kaupa
  ggplot(aes(y=filename,x=freq,color=searchterm))+
  geom_point(alpha=0.5)

words %>% 
  group_by(filename) %>% 
  count(word,sort=T) %>% 
  filter(str_detect(word,"w"))




keywords2 <- texts %>% 
  unnest_tokens(word,txt,to_lower=F) %>% 
  mutate(word=str_replace_all(word,"w","v")) %>% 
  anti_join(stopwords,"word") %>% 
  group_by(filename) %>% 
  filter(!str_detect(word,"[A-ZÕÄÖÜ]")) %>% 
  count(word,sort=T) %>% 
  bind_tf_idf(word,filename,n) %>% 
  arrange(desc(tf_idf)) %>% 
  filter(filename=="Juhan_Liiv_Noia_tutar.utf8") %>% 
  head(10)
keywords2

keywords <- texts %>% 
  unnest_tokens(word,txt,to_lower=F) %>% 
  mutate(word=str_replace_all(word,"w","v")) %>% 
  #anti_join(stopwords,"word") %>% 
  group_by(filename) %>% 
  filter(!str_detect(word,"[A-ZÕÄÖÜ]")) %>% 
  count(word,sort=T) %>% 
  bind_tf_idf(word,filename,n) %>% 
  filter(str_detect(filename,"Vambola")) %>% 
  arrange(desc(tf_idf)) %>% 
  head(10)

 filename                 word          n       tf   idf   tf_idf
   <chr>                    <chr>     <int>    <dbl> <dbl>    <dbl>
 1 Andres_Saal_Vambola.utf8 sõjamees     81 0.00120   1.61 0.00193 
 2 Andres_Saal_Vambola.utf8 vanake       64 0.000948  1.61 0.00153 
 3 Andres_Saal_Vambola.utf8 vaenlased    63 0.000933  1.46 0.00136 
 4 Andres_Saal_Vambola.utf8 neiuke       25 0.000370  3.40 0.00126 
 5 Andres_Saal_Vambola.utf8 vanema       80 0.00119   1.00 0.00119 
 6 Andres_Saal_Vambola.utf8 vaenlaste    65 0.000963  1.20 0.00116 
 7 Andres_Saal_Vambola.utf8 kalamees     32 0.000474  2.30 0.00109 
 8 Andres_Saal_Vambola.utf8 malev        21 0.000311  3.40 0.00106 
 9 Andres_Saal_Vambola.utf8 sõbranna     29 0.000430  2.30 0.000989
10 Andres_Saal_Vambola.utf8 sõjamehe     33 0.000489  2.01 0.000985