index.qmd

---
title: "What is this?"
author: "Lasse Hjorth Madsen"
date: today 
format: html
---

```{r load}
#| echo: false
profiles <- readRDS("data/research_profiles_2024-10-24.rds")
```

We've built a network of scientists and researchers active on Bluesky. The goal is to better understand how this group of users interact on the platform.

The network is built using an iterative algorithm that works like this: We start off with a list of hand-picked members of the scientific community, then expand the network step-by-step. This takes place by considering potential new members that:

  - are being followed by a substantial number from the existing network

  - have a bio description that indicates an affiliation with, or strong interest in, the scientific community
  
After a number of iterations, we arrived at a science-oriented network of `r format(nrow(profiles), big.mark = ",")` members (or 'actors') on Bluesky. Many, but not all, do research at a research institution. A few are science writers, independent researchers, or work at private organisations.

Once the network is established, we can compute centrality measures, like [betweenness centrality](https://en.wikipedia.org/wiki/Betweenness_centrality) or [PageRank](https://en.wikipedia.org/wiki/PageRank) to identify members that may be particularly influential.

Over at the list of [Influential Members of the Science Community](pages/big-list.qmd), we have a table with the top-100 members of the network, ranked by centrality. Feel free to explore the list, or play with the interactive [Network Visualisation](pages/big-plot.qmd).