text_in_R_Tbilisi.html

<!DOCTYPE html>
<html lang="" xml:lang="">
  <head>
    <title>Working with text in R</title>
    <meta charset="utf-8" />
    <meta name="author" content="Ella Kaye" />
    <script src="libs/header-attrs/header-attrs.js"></script>
    <link href="libs/remark-css/default.css" rel="stylesheet" />
    <link href="libs/remark-css/rladies.css" rel="stylesheet" />
    <link href="libs/remark-css/rladies-fonts.css" rel="stylesheet" />
    <script src="libs/htmlwidgets/htmlwidgets.js"></script>
    <link href="libs/str_view/str_view.css" rel="stylesheet" />
    <script src="libs/str_view-binding/str_view.js"></script>
  </head>
  <body>
    <textarea id="source">
class: left, middle, inverse, title-slide

# Working with text in R
### Ella Kaye
### July 4th, 2020, RLadies Tbilisi

---


# About me 

- Studying for a PhD in Statistics at the University of Warwick

- Previous degrees in Mathematics and Philosophy, History and Philosophy of Science, Mathematics Education, Secondary School Mathematics teaching

---

# Me and R

- Getting into R:

  - 2014: first R course
  - 2015: first taste of `ggplot2` 
  - 2016: first R package 
  - 2017: first community involvement (useR!2017, [Oxford R User Group](https://r-oxford.github.io), [rainbowR](https://rainbowr.slack.com), [@R_LGBTQ](https://twitter.com/R_LGBTQ))
  - 2020: first [TidyTuesday](https://github.com/rfordatascience/tidytuesday) contribution
  - 2021: first ...
  
---

class: center, inverse
background-image: url("edX_digital_humanities.png")
background-position: center
background-size: contain

---

class: center, inverse
background-image: url("Washington-and-Jefferson.jpg")
background-position: center
background-size: contain

---

# Command line

**Research question**: lengths of sentences in presidential speeches over the course of a presidency: terse and to the point at the beginning, long windered towards the end? Do they stay the same?

- Take word counts of each sentence and calculate the average word count per sentence per speech.

- Want table with year, month and sentence length recorded for each presidential speech. 

---

# [https://voyant-tools.org](https://voyant-tools.org)

![](Voyant.png)


---

# Libraries


```r
library(tidyverse) # especially `stringr`, `ggplot2`
library(RVerbalExpressions)
library(tidytext)
library(wordcloud)
library(ggwordcloud)
library(tidygraph)
library(ggraph)
```

---

# Inputting the text


```r
text &lt;- read_file("1790_01_08_Washington1.txt")
```

--

(show this)
---

# Regular expressions 

A sequence of characters that defines a search pattern.

--

.e$

--


```r
fruit &lt;- c("apple", "orange", "pear")
str_view(fruit, ".e$")
```

<div id="htmlwidget-1ad601b78d3e648cf8d6" style="width:960px;height:100%;" class="str_view html-widget"></div>
<script type="application/json" data-for="htmlwidget-1ad601b78d3e648cf8d6">{"x":{"html":"<ul>\n  <li>app<span class='match'>le<\/span><\/li>\n  <li>oran<span class='match'>ge<\/span><\/li>\n  <li>pear<\/li>\n<\/ul>"},"evals":[],"jsHooks":[]}</script>

---

# Basic `stringr`


```r
fruit &lt;- c("apple", "orange", "pear")
str_detect(fruit, ".e$")
```

```
## [1]  TRUE  TRUE FALSE
```

```r
str_extract(fruit, ".e$")
```

```
## [1] "le" "ge" NA
```

```r
str_subset(fruit, ".e$")
```

```
## [1] "apple"  "orange"
```


---

![](regular-expression-example.gif)

See [discussion](https://stackoverflow.com/questions/201323/how-to-validate-an-email-address-using-a-regular-expression/201378#201378)

---

# `RVerbalExpressions`

[https://github.com/VerbalExpressions/RVerbalExpressions](https://github.com/VerbalExpressions/RVerbalExpressions)


```r
span_rx &lt;- rx() %&gt;%
  rx_find("&lt;") %&gt;%
  rx_anything_but("&gt;") %&gt;%
  rx_find("&gt;")

span_rx
```

```
## [1] "(&lt;)([^&gt;]*)(&gt;)"
```

--

## Other Regular Expressions resources in R

- [rex](https://github.com/kevinushey/rex)
- [RegExplain](https://github.com/gadenbuie/regexplain)

---

# Cleaning the text


```r
text_clean &lt;- text %&gt;%
* str_remove_all(span_rx) %&gt;% # remove html tags
  str_remove_all("\\\\xe2\\\\x80\\\\x94") %&gt;% # remove unicode
  str_replace_all("&amp;#32", " ") %&gt;% # replace html space with space
* str_squish() %&gt;% # remove excess white space
  str_replace_all("-", " ") %&gt;% # remove hyphens (pros and cons)
  str_remove_all("[^[:alnum:][:space:].]") %&gt;% # remove punctuation except "."
  str_remove("This work is in the.*") # remove final sentence

write_file(text_clean, "1790_01_08_Washington1_clean.txt")
```

--

## `stringr` resources

- [`stringr` package site](https://stringr.tidyverse.org)
- [cheatsheet](https://resources.rstudio.com/rstudio-cheatsheets/stringr-cheat-sheet)
- [Strings chapter in R for Data Science](https://r4ds.had.co.nz/strings.html)

---

# Tidy Text

- The tidy text format is a table with one-token-per-row. 
- A token is a meaningful unit of text, such as a word, that we are interested in using for analysis
- Tokenization is the process of splitting text into tokens. 

When text is stored in this way, we can manipulate and plot it with other tidy tools, such as `dplyr`, `tidyr` and `ggplot2`.

--

The tidy text approach was developed by [Julia Silge](https://juliasilge.com) and [David Robinson](http://varianceexplained.org).

The main reference is the book [Text Mining with R: A tidy approach](https://www.tidytextmining.com). 

Tools for this approach are provided in the [`tidytext`](https://juliasilge.github.io/tidytext/) package.

---

# Tokenizing the Washington speech


```r
text_df &lt;- tibble(speech = text_clean) %&gt;%
* unnest_tokens(sentence, speech, token = "sentences")

text_df
```

```
## # A tibble: 28 x 1
##    sentence                                                                     
##    &lt;chr&gt;                                                                        
##  1 i embrace with great satisfaction the opportunity which now presents itself …
##  2 the recent accession of the important state of north carolina to the constit…
##  3 in resuming your consultations for the general good you cannot but derive en…
##  4 still further to realize their expectations and to secure the blessings whic…
##  5 among the many interesting objects which will engage your attention that of …
##  6 to be prepared for war is one of the most effectual means of preserving peac…
##  7 a free people ought not only to be armed but disciplined to which end a unif…
##  8 the proper establishment of the troops which may be deemed indispensible wil…
##  9 in the arrangements which may be made respecting it it will be of importance…
## 10 there was reason to hope that the pacific measures adopted with regard to ce…
## # … with 18 more rows
```

---

# Get sentence lengths

```r
text_df %&gt;%
  mutate(sentence_length = str_count(sentence, boundary("word"))) %&gt;%
  summarise(mean_sentence_length = mean(sentence_length))
```

```
## # A tibble: 1 x 1
##   mean_sentence_length
##                  &lt;dbl&gt;
## 1                 38.2
```

---

# Working with multiple text files


```r
washington_speeches &lt;- tibble(speech_name = c("1790_01_08_Washington1.txt",
                                              "1790_12_08_Washington2.txt",
                                              "1791_10_25_Washington3.txt",
                                              "1792_11_08_Washington4.txt"))

washington_speeches_df &lt;- washington_speeches %&gt;%
* mutate(speech = purrr::map_chr(speech_name, ~read_file(.x))) %&gt;%
  mutate(speech = str_remove_all(speech, span_rx)) %&gt;% # remove html tags
  mutate(speech = str_remove_all(speech, "\\\\xe2\\\\x80\\\\x94")) %&gt;% 
  mutate(speech = str_remove_all(speech, "&amp;#32")) %&gt;% # replace spaces
  mutate(speech = str_squish(speech)) %&gt;% # remove excess white space
  mutate(speech = str_replace_all(speech, "-", " ")) %&gt;% # remove hyphens 
  mutate(speech = str_remove_all(speech, "[^[:alnum:][:space:].]")) %&gt;%
  mutate(speech = str_remove(speech, "This work is in the.*")) %&gt;%
  mutate(speech_name = str_remove(speech_name, "[:digit:]\\.txt"))
```

---

# Split into sentences...


```r
washington_speeches_df %&gt;%
  unnest_tokens(sentence, speech, token = "sentences")
```

```
## # A tibble: 183 x 2
##    speech_name         sentence                                                 
##    &lt;chr&gt;               &lt;chr&gt;                                                    
##  1 1790_01_08_Washing… i embrace with great satisfaction the opportunity which …
##  2 1790_01_08_Washing… the recent accession of the important state of north car…
##  3 1790_01_08_Washing… in resuming your consultations for the general good you …
##  4 1790_01_08_Washing… still further torealize their expectations and to secure…
##  5 1790_01_08_Washing… among the many interesting objects which will engage you…
##  6 1790_01_08_Washing… to be prepared for war is one of the most effectual mean…
##  7 1790_01_08_Washing… a free people ought not only to be armed but disciplined…
##  8 1790_01_08_Washing… the proper establishment of the troops which may be deem…
##  9 1790_01_08_Washing… in the arrangements which may be made respecting it it w…
## 10 1790_01_08_Washing… there was reason to hope that the pacific measures adopt…
## # … with 173 more rows
```

---

# ...and get average sentence length


```r
washington_speeches_summary &lt;- washington_speeches_df %&gt;%
  unnest_tokens(sentence, speech, token = "sentences") %&gt;%
  mutate(sentence_length = str_count(sentence, boundary("word"))) %&gt;%
* group_by(speech_name) %&gt;%
  summarise(mean_sentence_length = mean(sentence_length)) %&gt;%
  separate(speech_name, sep = "_", into = c("Year", "Month", "Day", "President")) 

washington_speeches_summary
```

```
## # A tibble: 4 x 5
##   Year  Month Day   President  mean_sentence_length
##   &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt;                     &lt;dbl&gt;
## 1 1790  01    08    Washington                 38.1
## 2 1790  12    08    Washington                 36.0
## 3 1791  10    25    Washington                 39.7
## 4 1792  11    08    Washington                 35.6
```

--


```
## # A tibble: 4 x 5
##   Year  Month Day   President mean_sentence_length
##   &lt;chr&gt; &lt;chr&gt; &lt;chr&gt; &lt;chr&gt;                    &lt;dbl&gt;
## 1 1801  12    08    Jefferson                 37.2
## 2 1802  12    15    Jefferson                 36.7
## 3 1803  10    17    Jefferson                 49.3
## 4 1804  11    08    Jefferson                 44.7
```

---

# Word counts


```r
washington_by_word &lt;- washington_speeches_df %&gt;%
* unnest_tokens(word, speech, token = "words")
```

--


```r
washington_by_word %&gt;%
  count(word, sort = TRUE)
```

```
## # A tibble: 1,560 x 2
##    word      n
##    &lt;chr&gt; &lt;int&gt;
##  1 the     650
##  2 of      445
##  3 to      280
##  4 and     214
##  5 in      135
##  6 a       115
##  7 which   109
##  8 be      101
##  9 that     89
## 10 for      76
## # … with 1,550 more rows
```

---

# Stop words


```r
data(stop_words) # contained in `tidytext`

washington_words &lt;- washington_by_word %&gt;%
* anti_join(stop_words, by = "word")

washington_count &lt;- washington_words %&gt;%
  count(word, sort = TRUE)

washington_count
```

```
## # A tibble: 1,285 x 2
##    word           n
##    &lt;chr&gt;      &lt;int&gt;
##  1 united        28
##  2 public        25
##  3 citizens      17
##  4 government    17
##  5 measures      17
##  6 provision     17
##  7 proper        16
##  8 law           15
##  9 national      13
## 10 country       12
## # … with 1,275 more rows
```

---

class: center, middle

![Careful](https://media.giphy.com/media/WyDOpBN50SRy74GnqP/giphy.gif)

---

# What's missing?


```r
head(washington_count)
```

```
## # A tibble: 6 x 2
##   word           n
##   &lt;chr&gt;      &lt;int&gt;
## 1 united        28
## 2 public        25
## 3 citizens      17
## 4 government    17
## 5 measures      17
## 6 provision     17
```

--


```r
stop_words %&gt;%
  filter(word == "states")
```

```
## # A tibble: 1 x 2
##   word   lexicon
##   &lt;chr&gt;  &lt;chr&gt;  
## 1 states onix
```

---

# Try again...


```r
stop_words_modified &lt;- stop_words %&gt;%
  filter(word != "states")

washington_words &lt;- washington_by_word %&gt;%
  anti_join(stop_words_modified, by = "word")

washington_count &lt;- washington_words %&gt;%
  count(word, sort = TRUE)

washington_count
```

```
## # A tibble: 1,286 x 2
##    word           n
##    &lt;chr&gt;      &lt;int&gt;
##  1 states        29
##  2 united        28
##  3 public        25
##  4 citizens      17
##  5 government    17
##  6 measures      17
##  7 provision     17
##  8 proper        16
##  9 law           15
## 10 national      13
## # … with 1,276 more rows
```

---

# Visualising as bar chart

.pull-left[

```r
washington_count %&gt;%
  filter(n &gt; 9) %&gt;%
  mutate(word = reorder(word, n)) %&gt;%
  ggplot(aes(y = word, x = n)) + 
  geom_col() + 
  ylab(NULL) 
```
]

.pull-right[
![](text_in_R_Tbilisi_files/figure-html/plot-label-out-1.png)&lt;!-- --&gt;
]

---

# Visualising as `wordcloud`


```r
set.seed(2)
wordcloud(words = washington_count$word, freq = washington_count$n, 
          min.freq = 9, rot.per = 0.35, colors = brewer.pal(8, "YlGnBu"))
```

---

class: center, middle

![](text_in_R_Tbilisi_files/figure-html/wordcloud-out-1.png)&lt;!-- --&gt;

---

# Visualising as `ggwordcloud`


```r
set.seed(2)
washington_count %&gt;%
  filter(n &gt;= 9) %&gt;%
  mutate(angle = 90 * sample(c(0, 1), n(), replace = TRUE, prob = c(63, 35))) %&gt;%
  ggplot(aes(label = word, size = n, colour = n, angle = angle)) +
*   geom_text_wordcloud() +
    scale_radius(range = c(0, 15), limits = c(0, NA)) +
    theme_minimal() +
    scale_color_viridis_c(direction = -1, end = 0.9, begin = 0.1)
```

[`ggwordcloud` vignette](https://cran.r-project.org/web/packages/ggwordcloud/vignettes/ggwordcloud.html)
---

class: center, middle

![](text_in_R_Tbilisi_files/figure-html/ggwordcloud-out-1.png)&lt;!-- --&gt;

---

class: center, middle

![Connected](https://media.giphy.com/media/UrQ3YL7cwrMLLeuGix/source.gif)

---

# Bigrams


```r
washington_bigrams &lt;- washington_speeches_df %&gt;%
* unnest_tokens(bigram, speech, token = "ngrams", n = 2)

washington_bigrams
```

```
## # A tibble: 6,830 x 2
##    speech_name           bigram            
##    &lt;chr&gt;                 &lt;chr&gt;             
##  1 1790_01_08_Washington i embrace         
##  2 1790_01_08_Washington embrace with      
##  3 1790_01_08_Washington with great        
##  4 1790_01_08_Washington great satisfaction
##  5 1790_01_08_Washington satisfaction the  
##  6 1790_01_08_Washington the opportunity   
##  7 1790_01_08_Washington opportunity which 
##  8 1790_01_08_Washington which now         
##  9 1790_01_08_Washington now presents      
## 10 1790_01_08_Washington presents itself   
## # … with 6,820 more rows
```

---

# Separate out and remove stop words


```r
washington_bigrams &lt;- washington_bigrams %&gt;%
  separate(bigram, c("word1", "word2"), sep = " ") %&gt;%
  filter(!word1 %in% stop_words_modified$word) %&gt;%
  filter(!word2 %in% stop_words_modified$word) 
```


```r
washington_bigram_counts &lt;- washington_bigrams %&gt;%
  count(word1, word2, sort = TRUE)

washington_bigram_counts
```

```
## # A tibble: 483 x 3
##    word1    word2          n
##    &lt;chr&gt;    &lt;chr&gt;      &lt;int&gt;
##  1 united   states        27
##  2 fellow   citizens       6
##  3 post     office         5
##  4 post     roads          4
##  5 public   debt           4
##  6 national prosperity     3
##  7 3000000  florins        2
##  8 adequate provision      2
##  9 current  service        2
## 10 due      attention      2
## # … with 473 more rows
```

---

# Create a graph


```r
washington_bigram_graph &lt;- washington_bigram_counts %&gt;%
  filter(n &gt; 1) %&gt;%
* as_tbl_graph()
```

---

# The graph object


```r
washington_bigram_graph
```

```
## # A tbl_graph: 38 nodes and 23 edges
## #
## # A rooted forest with 15 trees
## #
## # Node Data: 38 x 1 (active)
##   name    
##   &lt;chr&gt;   
## 1 united  
## 2 fellow  
## 3 post    
## 4 public  
## 5 national
## 6 3000000 
## # … with 32 more rows
## #
## # Edge Data: 23 x 3
##    from    to     n
##   &lt;int&gt; &lt;int&gt; &lt;int&gt;
## 1     1    17    27
## 2     2    18     6
## 3     3    19     5
## # … with 20 more rows
```

---

# Plot the graph


```r
set.seed(1)
*ggraph(washington_bigram_graph, layout = "fr") +
  geom_edge_link() +
  geom_node_point() +
  geom_node_text(aes(label = name), vjust = 1, hjust = 1, repel = TRUE)
```

---

class: center, middle

![](text_in_R_Tbilisi_files/figure-html/first-graph-out-1.png)&lt;!-- --&gt;

---

# Improve the graph


```r
set.seed(1)

a &lt;- grid::arrow(type = "closed", length = unit(.1, "inches"))

ggraph(washington_bigram_graph, layout = "fr") +
* geom_edge_link(aes(edge_alpha = n), show.legend = FALSE, arrow = a,
*                end_cap = circle(.07, 'inches')) +
  geom_node_point(color = "lightblue", size = 5) +
  geom_node_text(aes(label = name), vjust = 1, hjust = 1, repel = TRUE) +
  theme_void()
```

---

class: center, middle

![](text_in_R_Tbilisi_files/figure-html/second-graph-out-1.png)&lt;!-- --&gt;

---

## Compare word frequencies in two sets of texts


```r
frequency &lt;- bind_rows(washington_words, jefferson_words) %&gt;%
  mutate(speech_name = str_extract_all(speech_name, "Washington|Jefferson", 
                                       simplify = TRUE)) %&gt;%
  rename(president = speech_name) %&gt;%
  count(president, word) %&gt;%
  group_by(president) %&gt;%
  mutate(proportion = n/sum(n)) %&gt;%
  select(-n) %&gt;%
  pivot_wider(names_from = "president", values_from = "proportion") 
```

--


```r
set.seed(234)
*slice_sample(frequency, n = 6) # dplyr 1.0.0
```

```
## # A tibble: 6 x 3
##   word         Jefferson Washington
##   &lt;chr&gt;            &lt;dbl&gt;      &lt;dbl&gt;
## 1 wise         NA          0.000794
## 2 intercourse   0.00140    0.00198 
## 3 engage       NA          0.00119 
## 4 revenue       0.00225    0.000397
## 5 commodities   0.000281  NA       
## 6 recollection NA          0.000397
```


---

# Plot


```r
ggplot(frequency, aes(x = Washington, y = Jefferson, 
                      color = abs(Washington - Jefferson))) +
  geom_abline(color = "gray40", lty = 2) +
  geom_jitter(alpha = 0.1, size = 2.5, width = 0.3, height = 0.3) +
  geom_text(aes(label = word), check_overlap = TRUE, vjust = 1.5) +
  scale_x_log10(labels = scales::percent_format()) +
  scale_y_log10(labels = scales::percent_format()) +
  scale_color_gradient(limits = c(0, 0.001), 
                       low = "darkslategray4", high = "gray75") +
  theme(legend.position="none") 
```

---

class: center, middle


![](text_in_R_Tbilisi_files/figure-html/frequency-plot-out-1.png)&lt;!-- --&gt;

---

class: center, middle


![Careful](https://media.giphy.com/media/dXLP6Hw8aMD9VqwicP/source.gif)

---

# Sentiment: AFINN

Score -5 (most negative) to 5 (most positive)


```r
get_sentiments("afinn") 
```

```
## # A tibble: 2,477 x 2
##    word       value
##    &lt;chr&gt;      &lt;dbl&gt;
##  1 abandon       -2
##  2 abandoned     -2
##  3 abandons      -2
##  4 abducted      -2
##  5 abduction     -2
##  6 abductions    -2
##  7 abhor         -3
##  8 abhorred      -3
##  9 abhorrent     -3
## 10 abhors        -3
## # … with 2,467 more rows
```

---

# Sentiment: bing

binary positive/negative


```r
get_sentiments("bing") 
```

```
## # A tibble: 6,786 x 2
##    word        sentiment
##    &lt;chr&gt;       &lt;chr&gt;    
##  1 2-faces     negative 
##  2 abnormal    negative 
##  3 abolish     negative 
##  4 abominable  negative 
##  5 abominably  negative 
##  6 abominate   negative 
##  7 abomination negative 
##  8 abort       negative 
##  9 aborted     negative 
## 10 aborts      negative 
## # … with 6,776 more rows
```

---

# Sentiment: nrc

categorizes words in a binary fashion (“yes”/“no”) into categories of positive, negative, anger, anticipation, disgust, fear, joy, sadness, surprise, and trust


```r
get_sentiments("nrc") 
```

```
## # A tibble: 13,901 x 2
##    word        sentiment
##    &lt;chr&gt;       &lt;chr&gt;    
##  1 abacus      trust    
##  2 abandon     fear     
##  3 abandon     negative 
##  4 abandon     sadness  
##  5 abandoned   anger    
##  6 abandoned   fear     
##  7 abandoned   negative 
##  8 abandoned   sadness  
##  9 abandonment anger    
## 10 abandonment fear     
## # … with 13,891 more rows
```

---

# Notes on sentiments

- Note that all these lexicons are on unigrams, so don't pick up things like 'no good' as negative

--

- Constructed via either crowdsourcing (using, for example, Amazon Mechanical Turk) or by the labor of one of the authors

--

- Validated using some combination of crowdsourcing again, restaurant or movie reviews, or Twitter data

--

- Be hesitant about applying these sentiment lexicons to styles of text dramatically different from what they were validated on, such as presidential speeches from 230 years ago!

--

- There are some other domain-specific sentiment lexicons available, constructed to be used with text from a specific content area


---

# Washington sentiment: code


```r
washington_words %&gt;%
* inner_join(get_sentiments("bing"), by = "word") %&gt;%
  count(word, sentiment, sort = TRUE) %&gt;%
  ungroup() %&gt;%
  group_by(sentiment) %&gt;%
* slice_max(order_by = n, n = 5) %&gt;% # dplyr 1.0.0
  ungroup() %&gt;%
  mutate(word = reorder(word, n)) %&gt;%
* ggplot(aes(y = word, x = n, fill = sentiment)) + # ggplot2 3.3.0
    geom_col(show.legend = FALSE) +
    facet_wrap(~sentiment, scales = "free_y") +
    labs(x = "Contribution to sentiment",
       y = NULL) 
```

---

# Washington sentiment: plot

&lt;img src="text_in_R_Tbilisi_files/figure-html/wash-sent-out-1.png" style="display: block; margin: auto;" /&gt;

---

class: center

background-image: url("pride-prej-sent.png")
background-position: center
background-size: contain

---

class: center

background-image: url("2020_19_AnimalCrossing_alt_50.png")
background-position: center
background-size: contain

---

# Some other resources 

- [Code for graph on previous slide](https://github.com/Z3tt/TidyTuesday/blob/master/R/2020_19_AnimalCrossing.Rmd)
- [readtext.quanteda.io](https://readtext.quanteda.io) - import and handling for plain and formatted text files
- [pdftools](https://docs.ropensci.org/pdftools/) - extracting text and metadata from pdf files in R
- [tesseract](https://cran.r-project.org/web/packages/tesseract/vignettes/intro.html) - optical character recognition in R
- [quantedo.io](http://quanteda.io) - managing and analyzing texts; apply natural language processing to texts

[@rivaquiroga](https://twitter.com/rivaquiroga) (via [@WeAreRLadies](https://twitter.com/WeAreRLadies))

---
class: center, middle, inverse

## Please ask me questions!


## I'd love to hear from you:
[E.Kaye.1@warwick.ac.uk](mailto:E.Kaye.1@warwick.ac.uk)

[@ellamkaye](https://twitter.com/ellamkaye)

[ellakaye.rbind.io](https://ellakaye.rbind.io)

[github.com/EllaKaye](https://github.com/ellakaye)
    </textarea>
<style data-target="print-only">@media screen {.remark-slide-container{display:block;}.remark-slide-scaler{box-shadow:none;}}</style>
<script src="https://remarkjs.com/downloads/remark-latest.min.js"></script>
<script>var slideshow = remark.create({
"highlightStyle": "github",
"highlightLines": true,
"countIncrementalSlides": false
});
if (window.HTMLWidgets) slideshow.on('afterShowSlide', function (slide) {
  window.dispatchEvent(new Event('resize'));
});
(function(d) {
  var s = d.createElement("style"), r = d.querySelector(".remark-slide-scaler");
  if (!r) return;
  s.type = "text/css"; s.innerHTML = "@page {size: " + r.style.width + " " + r.style.height +"; }";
  d.head.appendChild(s);
})(document);

(function(d) {
  var el = d.getElementsByClassName("remark-slides-area");
  if (!el) return;
  var slide, slides = slideshow.getSlides(), els = el[0].children;
  for (var i = 1; i < slides.length; i++) {
    slide = slides[i];
    if (slide.properties.continued === "true" || slide.properties.count === "false") {
      els[i - 1].className += ' has-continuation';
    }
  }
  var s = d.createElement("style");
  s.type = "text/css"; s.innerHTML = "@media print { .has-continuation { display: none; } }";
  d.head.appendChild(s);
})(document);
// delete the temporary CSS (for displaying all slides initially) when the user
// starts to view slides
(function() {
  var deleted = false;
  slideshow.on('beforeShowSlide', function(slide) {
    if (deleted) return;
    var sheets = document.styleSheets, node;
    for (var i = 0; i < sheets.length; i++) {
      node = sheets[i].ownerNode;
      if (node.dataset["target"] !== "print-only") continue;
      node.parentNode.removeChild(node);
    }
    deleted = true;
  });
})();
(function() {
  "use strict"
  // Replace <script> tags in slides area to make them executable
  var scripts = document.querySelectorAll(
    '.remark-slides-area .remark-slide-container script'
  );
  if (!scripts.length) return;
  for (var i = 0; i < scripts.length; i++) {
    var s = document.createElement('script');
    var code = document.createTextNode(scripts[i].textContent);
    s.appendChild(code);
    var scriptAttrs = scripts[i].attributes;
    for (var j = 0; j < scriptAttrs.length; j++) {
      s.setAttribute(scriptAttrs[j].name, scriptAttrs[j].value);
    }
    scripts[i].parentElement.replaceChild(s, scripts[i]);
  }
})();
(function() {
  var links = document.getElementsByTagName('a');
  for (var i = 0; i < links.length; i++) {
    if (/^(https?:)?\/\//.test(links[i].getAttribute('href'))) {
      links[i].target = '_blank';
    }
  }
})();
// adds .remark-code-has-line-highlighted class to <pre> parent elements
// of code chunks containing highlighted lines with class .remark-code-line-highlighted
(function(d) {
  const hlines = d.querySelectorAll('.remark-code-line-highlighted');
  const preParents = [];
  const findPreParent = function(line, p = 0) {
    if (p > 1) return null; // traverse up no further than grandparent
    const el = line.parentElement;
    return el.tagName === "PRE" ? el : findPreParent(el, ++p);
  };

  for (let line of hlines) {
    let pre = findPreParent(line);
    if (pre && !preParents.includes(pre)) preParents.push(pre);
  }
  preParents.forEach(p => p.classList.add("remark-code-has-line-highlighted"));
})(document);</script>

<script>
slideshow._releaseMath = function(el) {
  var i, text, code, codes = el.getElementsByTagName('code');
  for (i = 0; i < codes.length;) {
    code = codes[i];
    if (code.parentNode.tagName !== 'PRE' && code.childElementCount === 0) {
      text = code.textContent;
      if (/^\\\((.|\s)+\\\)$/.test(text) || /^\\\[(.|\s)+\\\]$/.test(text) ||
          /^\$\$(.|\s)+\$\$$/.test(text) ||
          /^\\begin\{([^}]+)\}(.|\s)+\\end\{[^}]+\}$/.test(text)) {
        code.outerHTML = code.innerHTML;  // remove <code></code>
        continue;
      }
    }
    i++;
  }
};
slideshow._releaseMath(document);
</script>
<!-- dynamically load mathjax for compatibility with self-contained -->
<script>
(function () {
  var script = document.createElement('script');
  script.type = 'text/javascript';
  script.src  = 'https://mathjax.rstudio.com/latest/MathJax.js?config=TeX-MML-AM_CHTML';
  if (location.protocol !== 'file:' && /^https?:/.test(script.src))
    script.src  = script.src.replace(/^https?:/, '');
  document.getElementsByTagName('head')[0].appendChild(script);
})();
</script>
  </body>
</html>