Polarized and nonpolarized Twitter networks from the 2019 Finnish Parliamentary Elections This dataset includes 183 Twitter retweet networks collected during the 2019 Finnish Parliamentary Elections. The first 150 networks are built around single hashtags, such as #police, #nature, and #immigration. The remaining 33 networks are constructed using a combination of hashtags focused on specific topics like climate change and economic policy. Each filename consists of two parts: the first part indicates whether the network is based on a single hashtag (in lowercase) or a set of hashtags (in uppercase). The second part represents the tweet period. "p1" corresponds to the pre-election period (March 1 to April 14). "p2" corresponds to the inter-election period (April 15 to May 26). "p3" corresponds to the post-election period (May 27 to July 31). The nodes in the networks represent anonymized Twitter accounts, and directed ties indicate retweet endorsements on specific topics. Each file contains three columns: retweeter, retweeted, and weight. Please see the references for more details. Network labels, whether they are labeled as controversial, and whether they are based on single or multiple hashtags, can be found in the "networks_info.csv" file. Importantly, the dataset does not contain any identifying information or original raw data from the Twitter platform. Anonymization was achieved by shuffling the order of unique nodes across all networks and assigning each node a new identifier (ID). These new IDs were then applied to the edgelists to obtain the anonymized version. Kindly ensure to reference the original article(s) when utilizing this dataset. Chen, T. H. Y., Salloum, A., Gronow, A., Ylä-Anttila, T., & Kivelä, M. (2021). Polarization of climate politics results from partisan sorting: Evidence from Finnish Twittersphere. Global Environmental Change, 71, 102348. https://doi.org/10.1016/j.gloenvcha.2021.102348 Salloum, A., Chen, T. H. Y., & Kivelä, M. (2022). Separating polarization from noise: comparison and normalization of structural polarization measures. Proceedings of the ACM on human-computer interaction, 6(CSCW1), 1-33. https://doi.org/10.1145/3512962

{"references": ["Chen, T. H. Y., Salloum, A., Gronow, A., Yl\u00e4-Anttila, T., & Kivel\u00e4, M. (2021). Polarization of climate politics results from partisan sorting: Evidence from Finnish Twittersphere. Global Environmental Change, 71, 102348. https://doi.org/10.1016/j.gloenvcha.2021.102348", "Salloum, A., Chen, T. H. Y., & Kivel\u00e4, M. (2022). Separating polarization from noise: comparison and normalization of structural polarization measures. Proceedings of the ACM on human-computer interaction, 6(CSCW1), 1-33. https://doi.org/10.1145/3512962"]}

Please use Version 2 data (Updated 12.10.2023). Changes: + More detailed description added + Appropriate anonymization of nodes that is consistent across all the networks, i.e. same Twitter account will have identical ID in all the networks they belong to.