Supplementary Materials

New online ecology of adversarial aggregates: ISIS and beyond

N. F. Johnson, M. Zheng, Y. Vorobyeva, A. Gabriel, H. Qi, N. Velasquez, P. Manrique, D. Johnson, E. Restrepo, C. Song, S. Wuchty

Materials/Methods, Supplementary Text, Tables, Figures, and/or References

Download Supplement
  • Materials and Methods
  • Supplementary Text
  • Figs. S1 to S12
  • Tables S1 to S3
  • References
Data S1
The datafile can be easily opened using any spreadsheet software. It contains data used to produce curves in the graphs shown in the paper. The sheets in this file are as follows and have the following content: Fig.1. Bipartite network of aggregates and followers. Column B corresponds to the anonymized identity of follower and Column C the anonymized identity of aggregate. Follower and aggregate in the same row are connected; Fig.2(b) List of Facebook aggregates in Brazil. Column A contains an anonymized version of the aggregate’s name (name available from authors) and Column B the corresponding date of creation; Fig.2(c) Data-points of escalation parameter with the corresponding date for pro-ISIS aggregates; Fig.2(d) Data-points of escalation parameter with the corresponding date for Brazil aggregates; Fig.3(a) Data-points of number of follows (from Column C to AD) as a function of time located in Column B; Fig.3(c) Data-points of the empirical distribution of aggregate sizes; Fig.4(a) Name change adaptation. Column B contains time. Columns C to AI aggregate identities (anonymized) followed by integer numbers from 0 to 6. Number 0 means that the aggregate does not exist at that timestep while numbers 1-5 corresponds to variations on aggregate’s name; Fig.4(b) Invisibility adaptation. Column B contains time. Columns C to P aggregate identities (anonymized) followed by integer numbers between 0 and 2. Number 0 means that the aggregate does not exist at that timestep. Numbers 1 and 2 means that the aggregate at that timestep is visible and invisible, respectively; Fig.4(c) Reincarnation adaptation. Column B contains time. Columns C to J aggregate followed by integer numbers between 0 and 5. Number 0 means that the aggregate does not exist on that timestep. Number 1 represents the original aggregate. Numbers from 2 to 5 are revivals of the original aggregate that started with a significant percentage of follows that came from the original aggregate; Fig.S4. Civil unrest events in Brazil in 2013. Part A contains a table with information about Brazil protests including state, city, geolocation (longitude and latitude), date and number of protestors. Part B contains the consolidated number of events for a particular date; Fig.S6. Part A contains a list of Facebook aggregates in Venezuela. Column B contains the aggregate’s name (anonymized) and Column C the corresponding date of creation. Part B contains the data-points of escalation parameter with the corresponding date for aggregates in Venezuela; Fig.S5. Data-points of escalation parameter from Facebook data (Column C) and from event data (Column F) with the corresponding dates (Column B and E, respectively) for Brazil aggregates; Fig.S1. Civil unrest events and Google Trends (GT) volumes in Chile. GT volumes of search term 'Protesta' (Column E) and 'Huelga' (Column F) for a particular date (Columns B-D). Number of civil unrest events events (Column K) for a particular date (Column H-J)