[Dataset] Instagram Follower Networks, Suspicious Spikes, And Possible Coordinated Activity

With parliamentary elections approaching in Serbia, numerous Instagram accounts critical of the ruling government have experienced sudden waves of suspicious followers, sometimes thousands within just a few hours.

We’ve collected around 1.3 million follower records and published them as an open dataset on Hugging Face:

https://huggingface.co/datasets/gavrilo/botijada

The dataset includes follow timestamps and pseudonymized account IDs that remain consistent across different profiles.

There are quite a few interesting directions to explore here: temporal patterns, shared followers between accounts, network structures, clustering, anomaly detection, and potentially signs of coordinated activity.

I’m sharing this mostly because I think there’s a lot of room to experiment with the data, especially if you’re into graph analysis, data visualization, or finding interesting patterns in messy real-world datasets.

Feel free to play around with it, build something, or share whatever you discover. I’d be genuinely interested in seeing different approaches and interpretations.