ESA100tweets.Rmd

---
title: A summary of the tweets generated at the 2015 Ecological Society of America
  meeting (#ESA100)
author: "E.J. Rollinson"
output:
  html_document:
    toc: yes
  pdf_document:
    toc: yes
---  

```{r, echo=FALSE, results='hide', message=FALSE}
dt_tweets <- read.csv("ESA100tweets.csv", stringsAsFactors = FALSE)
dt_tweets$created <- as.Date(dt_tweets$created)

library(dplyr)
library(ggplot2)
library(knitr)
library(tidyr)
library(wesanderson)
```

## About this document

The code generating this document was originally developed by [Francois Michonneau](https://github.com/fmichonneau) ([fmic_](https://twitter.com/fmic_) on Twitter) for the 2015 Evolution meeting and can be found [here](https://github.com/fmichonneau/evol2015-tweets). 

It has since been [adapted](https://github.com/jlehtoma/iccb2015-tweets/blob/gh-pages/index.Rmd) for ICCB-ECCB 2015 by Joona Lehtomaki ([jlehtoma](https://twitter.com/jlehtoma) on Twitter).

Thanks also to Emily Sessa ([ebsessa](http://twitter.com/ebsessa) on Twitter) for reproducing this code for the 2015 Botany meeting, bringing it to my attention.

Here, the code is adapted for the 2015 Ecological Society of America Meeting (ESA100) in Baltimore, MD, by Emily J. Rollinson ([ejrollinson](https://twitter.com/ejrollinson), prev. katteken, on Twitter). 
 

Tweets using the hashtag #ESA100 were aggregated from Twitter using the R package [twitteR](https://cran.r-project.org/web/packages/twitteR/index.html) and the Twitter API.  The aggregated tweets are [available on Figshare](http://dx.doi.org/10.6084/m9.figshare.1512350).

This document was generated using RMarkdown, and the source is [available on GitHub](https://github.com/erollinson/ESA100-tweets).

This document (and associated code) is released under a CC0 licence.


## Basic summary

Of the most recent 10k tweets tagged #ESA100 (As of 9:00 PM 8/15/2015):

|Description | n |
|------------|---|
|Total of original tweets (no retweets): | `r sum(!dt_tweets$isRetweet)`|
|Number of users who tweeted: | `r length(unique(dt_tweets$screenName))`|

| ---
## The 5 most favorited tweets

```{r top-fav, echo=FALSE, results='asis'}
top_fav <- dt_tweets %>%
  filter(!isRetweet) %>%
  arrange(desc(favoriteCount)) %>%
  slice(1:5)


render_tweet <- function(dt, row) {
    screen_name <- dt[i, "screenName"]
    id <- format(dt[i, "id"], scientific = FALSE)
    txt <- dt[i, "text"]
    created <- format(dt[i, "created"], "%Y-%m-%d")
    n_fav <- dt[i, "favoriteCount"]
    n_retweets <- dt[i, "retweetCount"]
    cat("<blockquote class=\"twitter-tweet\" lang=\"en\"> \n",
        "<p lang=\"en\" dir=\"ltr\">",
        txt,
        "</p>&mdash; ",
        "<a href=\"https://twitter.com/", screen_name, "\">", screen_name, "</a>", "&nbsp;|&nbsp;",
        "<a href=\"https://twitter.com/",
        screen_name, "/status/", id, "\"> ", created, "</a> &nbsp;|&nbsp;",
        n_retweets, " retweets, ",  n_fav, " favorites. </blockquote>",
        "\n \n",
        sep = "")
}

for (i in seq_len(nrow(top_fav))) {
    render_tweet(top_fav, i)
}


```


## The 5 most retweeted tweets

```{r top-rt, echo=FALSE, results='asis'}
top_rt <- dt_tweets %>%
  filter(!isRetweet) %>%
  arrange(desc(retweetCount)) %>%
  slice(1:5)

for (i in seq_len(nrow(top_fav))) {
    render_tweet(top_rt, i)
}

```

## Top tweeters

All generated tweets (including retweets)

```{r top-users-all, echo=FALSE, fig.height=10}
top_users <- dt_tweets %>% group_by(screenName) %>%
  summarize(total_tweets = n(),
            Retweet = sum(isRetweet),
            Original = sum(!isRetweet)) %>%
  arrange(desc(total_tweets)) %>%
  slice(1:50) %>%
  gather(type, n_tweets, -screenName, -total_tweets)

top_users$screenName <- reorder(top_users$screenName,
                                top_users$total_tweets,
                                function(x) sum(x))

ggplot(top_users) + geom_bar(aes(x = screenName, y = n_tweets, fill = type),
                             stat = "identity") +
  ylab("Number of tweets") + xlab("User") +
  coord_flip() +
  scale_fill_manual(values = wes_palette("Zissou")[c(1, 3)]) +
  theme(axis.text = element_text(size = 12),
        legend.text = element_text(size = 12))
```

Only for original tweets (retweets excluded)

```{r, top-users-orig, echo=FALSE, fig.height=10}
top_orig_users <- dt_tweets %>% group_by(screenName) %>%
  summarize(total_tweets = n(),
            Retweet = sum(isRetweet),
            Original = sum(!isRetweet)) %>%
  arrange(desc(Original)) %>%
  slice(1:50)

top_orig_users$screenName <- reorder(top_orig_users$screenName,
                                     top_orig_users$Original,
                                     function(x) sum(x))

## png(file = "top_users2.png", width = 800, height = 800)
ggplot(top_orig_users) + geom_bar(aes(x = screenName, y = Original), stat = "identity",
                                  fill = wes_palette("Zissou", 1)) +
  ylab("Number of tweets") + xlab("User") +
  coord_flip() +
  theme(axis.text = element_text(size = 12),
        legend.text = element_text(size = 12))
## dev.off()

```


## Most favorited/retweeted users

The figures below only include users who tweeted 5+ times, and don't include retweets.

### Number of favorites received by users

```{r, fig.height=10, echo=FALSE}
impact <- dt_tweets %>% filter(!isRetweet) %>%
  group_by(screenName) %>%
  summarize(n_tweets = n(),
            n_fav = sum(favoriteCount),
            n_rt =  sum(retweetCount),
            mean_fav = mean(favoriteCount),
            mean_rt = mean(retweetCount)) %>%
  filter(n_tweets >=  5)

### Most favorited
most_fav <- impact %>%
  arrange(desc(n_fav)) %>%
  slice(1:50)

most_fav$screenName <- reorder(most_fav$screenName,
                               most_fav$n_fav,
                               sort)

ggplot(most_fav) + geom_bar(aes(x = screenName, y = n_fav),
                            stat = "identity", fill = wes_palette("Zissou")[2]) +
  coord_flip() + xlab("User") + ylab("Total number of favorites") +
  theme(axis.text = element_text(size = 12),
        legend.text = element_text(size = 12))
```

### Number of retweets received by users

```{r, fig.height=10, echo=FALSE}
## Most retweeted

most_rt <- impact %>%
  arrange(desc(n_rt)) %>%
  slice(1:50)

most_rt$screenName <- reorder(most_rt$screenName,
                              most_rt$n_rt,
                              sort)

ggplot(most_rt) + geom_bar(aes(x = screenName, y = n_rt),
                           stat = "identity", fill = wes_palette("Zissou")[5]) +
  coord_flip() + xlab("User") + ylab("Total number of retweets") +
  theme(axis.text = element_text(size = 12),
        legend.text = element_text(size = 12))
```

### Mean numbers of favorites received

```{r, fig.height=10, echo=FALSE}

### Mean favorites

hi_mean_fav <- impact %>%
  arrange(desc(mean_fav)) %>%
  slice(1:50)

hi_mean_fav$screenName <- reorder(hi_mean_fav$screenName,
                                  hi_mean_fav$mean_fav,
                                  sort)

ggplot(hi_mean_fav) + geom_bar(aes(x = screenName, y = mean_fav),
                           stat = "identity", fill = wes_palette("Zissou")[2]) +
  coord_flip() + xlab("User") + ylab("Number of favorites / tweets") +
  theme(axis.text = element_text(size = 12),
        legend.text = element_text(size = 12))

```

### Mean numbers of retweets received

```{r, fig.height=10, echo=FALSE}

### Mean retweets

hi_mean_rt <- impact %>%
  arrange(desc(mean_rt)) %>%
  slice(1:50)

hi_mean_rt$screenName <- reorder(hi_mean_rt$screenName,
                                 hi_mean_rt$mean_rt,
                                 sort)

ggplot(hi_mean_rt) + geom_bar(aes(x = screenName, y = mean_rt),
                           stat = "identity", fill = wes_palette("Zissou")[5]) +
  coord_flip() + xlab("User") + ylab("Number of retweets / tweets") +
  theme(axis.text = element_text(size = 12),
        legend.text = element_text(size = 12))


```

## Word cloud

The top 100 words among the original tweets.

```{r word-cloud, echo=FALSE, message=FALSE}
library(wordcloud)

pal <- wes_palette("Darjeeling", 8, type = "continuous") #brewer.pal(8, "Dark2")

dt_tweets %>%
  filter(!isRetweet) %>%
  .$text %>% paste(collapse = "") %>%
  gsub("(@|\\#)\\w+", "", .) %>%  ## remove mentions/hashtags
  gsub("https?\\:\\/\\/\\w+\\.\\w+(\\/\\w+)*", "", .) %>% ## remove urls
  gsub("\\bthe\\b", "", .) %>% ## remove the
  gsub("amp", "", .) %>%  ## remove &
  gsub("\\bspp\\b", "species", .) %>% ## replace spp by species
  iconv(., from = "latin1", to = "UTF-8", sub = "") %>% ## remove emojis
  wordcloud(max.words = 100, colors = pal, random.order = FALSE, scale = c(3, .7))

```

-----

<p xmlns:dct="http://purl.org/dc/terms/" xmlns:vcard="http://www.w3.org/2001/vcard-rdf/3.0#">
  <a rel="license"
     href="http://creativecommons.org/publicdomain/zero/1.0/">
    <img src="http://i.creativecommons.org/p/zero/1.0/88x31.png" style="border-style: none;" alt="CC0" />
  </a>
  <br />
  To the extent possible under law,
  <a rel="dct:publisher"
     href="https://github.com/fmichonneau/evol2015-tweets">
    <span property="dct:title">Emily J. Rollinson</span></a>
  has waived all copyright and related or neighboring rights to
  <span property="dct:title">Summary of tweets at the 2015 Ecological Society of America conference</span>.
This work is published from:
<span property="vcard:Country" datatype="dct:ISO3166"
      content="US" about="https://github.com/erollinson/ESA100-tweets">
  United States</span>.
</p>