twitter.Rmd

---
title: "Example: Social media analysis (Twitter)"
author: "Kohei Watanabe"
output: 
  html_document:
    toc: true
---

```{r, echo = FALSE}
knitr::opts_chunk$set(
  collapse = FALSE,
  comment = "##"
)
```

Using **quanteda**'s `fcm()` and `textplot_network()`, you can perform visual analysis of social media posts in terms of co-occurrences of hashtags or usernames in a few steps. The dataset for this example contains only 10,000 Twitter posts, but you can easily analyse more than one million posts on your laptop computer.

```{r, message = FALSE}
library(quanteda)
```

## Load sample data
```{r}
load("data/data_corpus_tweets.rda")
```

## Construct a document-feature matrix of Twitter posts
```{r}
dfmat_tweets <- tokens(data_corpus_tweets, remove_punct = TRUE) |>
    dfm()
head(dfmat_tweets)
```

# Hashtags

## Extract most common hashtags
```{r}
dfmat_tag <- dfm_select(dfmat_tweets, pattern = "#*")
toptag <- names(topfeatures(dfmat_tag, 50))
head(toptag)
```

## Construct feature-occurrence matrix of hashtags
```{r}
library("quanteda.textplots")
fcmat_tag <- fcm(dfmat_tag)
head(fcmat_tag)
fcmat_topgat <- fcm_select(fcmat_tag, pattern = toptag)
textplot_network(fcmat_topgat, min_freq = 0.1, edge_alpha = 0.8, edge_size = 5)
```

# Usernames

## Extract most frequently mentioned usernames 
```{r}
dfmtat_users <- dfm_select(dfmat_tweets, pattern = "@*")
topuser <- names(topfeatures(dfmtat_users, 50))
head(topuser)
```

## Construct feature-occurrence matrix of usernames
```{r}
fcmat_users <- fcm(dfmtat_users)
head(fcmat_users)
fcmat_users <- fcm_select(fcmat_users, pattern = topuser)
textplot_network(fcmat_users, min_freq = 0.1, edge_color = "orange", edge_alpha = 0.8, edge_size = 5)
```