Planned intervention: On Wednesday April 3rd 05:30 UTC Zenodo will be unavailable for up to 2-10 minutes to perform a storage cluster upgrade.
Published December 13, 2021 | Version 1.0
Dataset Open

Twitter hashtags time series used in the paper "Universality, criticality and complexity of information propagation in social media"

  • 1. Center for Complex Networks and Systems Research, Luddy School of Informatics, Computing, and Engineering, Indiana University, Bloomington, Indiana 47408, USA
  • 2. Istituto dei Sistemi Complessi (ISC-CNR), Via dei Taurini 19, I-00185 Roma, Italy.
  • 3. Center for Complex Networks and Systems Research, Luddy School of Informatics, Computing, and Engineering, Indiana University, Bloomington, Indiana 47408, USA AND Centro Fermi, Via Panisperna 89 A, Roma, Italy.

Description

These files contain the time series and the associated hashtags we obtained by sampling 
Twitter for our paper "Universality, criticality and complexity of information 
propagation on social media". The analysis is reported in 

https://www.nature.com/articles/s41467-022-28964-8

Please acknowledge the use of these data by citing the paper above.


#################################
#################################


DATA ORGANIZATION


We created a single zip file with all the time series and a single zip file 
with all the hashtags. There is a one-to-one correspondence between lines in 
the two files. 

#################################
#################################


FILES CONTENT


As stated, here is a one-to-one correspondence between lines in the time series 
file and lines in the hashtags file, i.e., the hashtag 
stored in line X is the hashtag of the time series stored in line X. 
Time series are stored as follows:

Ka t1 t2 t3 \n
Kb t1 t2 t3 t4 t5 \n
.
.
.
Kn t1 t2 \n

where:

Ka, Kb,..., Kn is an integer specifying the number of events that compose 
the time series a, b,..., n respectively. In the example above 
we would have Ka=3, Kb=5, Kn=2.

t1 t2 ... is the time series, i.e., a sequence of chronologically ordered 
interevent times. The last interevent time, in our implementation, represents 
the distance between the end of the temporal window and the last event time. 
It thus does not represent an event. As stated in the Supplemental Material of 
our paper, the temporal window ranges from 2019, October 1st to 2019, November 30th.

Files

twitter_hashtags.txt.zip

Files (1.2 GB)

Name Size Download all
md5:8fbfbbaf28ee78af44258e10d5a59bee
173.8 MB Preview Download
md5:ca5e950cf3bbdb6e3ced506a23eea67a
1.1 GB Preview Download
md5:679ec526449ca89909b9adf2e57d94a1
212.6 kB Preview Download

Additional details

Funding

NRT: Interdisciplinary Training in Complex Networks and Systems 1735095
National Science Foundation
CAREER: Network Theory of Critical Interdependent Infrastructures 1552487
National Science Foundation