10 Twitter Networks on 5 Topics in Finnish Twittersphere (2019 & 2023)

Description

This dataset contains 10 undirected networks derived from Twitter data (API v1 & v2), covering five distinct political topics for two different parliamentary election years:

• Topics: Climate, Immigration, Social Security, Economy, Education

• Years: 2019 and 2023

This deposit includes the final, processed network files in .graphml format. All nodes (users) across all 10 networks have been relabeled with a single, consistent, global integer ID (from 0 to N-1, where N is the total number of unique users). This anonymizes the data and allows for direct comparison and tracking of users across different networks and time periods.

Data Collection

See section 4.1 in Salloum, Chen & Kivelä (2025) Anatomy of elite and mass polarization in social networks.

Files

This deposit contains the following files:

• climate_19.graphml

• climate_23.graphml

• immigration_19.graphml

• immigration_23.graphml

• social_19.graphml

• social_23.graphml

• economy_19.graphml

• economy_23.graphml

• education_19.graphml

• education_23.graphml

Data Dictionary

Relabeled GraphML Files (*.graphml)

These files can be opened by any network analysis software that supports the GraphML format (e.g., Gephi, Python networkx, R igraph).

• Graph Type: Undirected (Graph)

• Nodes: Represent individual users.

  • id (Integer): The new, global, persistent integer ID for the user (from 0 to N-1).

  • group (String): The community or group label (A or B) assigned to the node.

  • hierarchy (String): The hierarchical status (CORE or PERIPHERY) assigned to the node.

• Edges: Represent a retweet with no added comment (quotations) between users. 

How to Use

The .graphml files can be loaded directly for analysis.

Python (networkx) Example:

import networkx as nx

# Load a network
G = nx.read_graphml("./climate_23.graphml")

Cite As

Kindly ensure to reference the original article when utilizing this dataset: Salloum, Chen & Kivelä (2025) Anatomy of elite and mass polarization in social networks.