Diurnal Changes Datasets

This web page documents our datasets about detecting Diurnal Changes in address blocks via Internet measurements. One cause of such changes is Covid Work-from-Home (WFH). Our datasets are available upon request.

See also: original Diurnal datasets, IP address accumulation datsets.

What data do you have? We have global, daily data for 2020 January to June.

What is in the data? We document the format of outage datasets.

Relevant Publications

These datasets have been used in the following publications:

  • peer-reviewed paper about detecting changes in human behavior: [1]
  • poster [2] and technical report [3] about Covid-WFH detection
  • peer-reviewed paper about visualizing Covid-WFH detection: [4]
  • Xiao Song, Guillermo Baltra and John Heidemann 2023. Inferring Changes in Daily Human Activity from Internet Response. Proceedings of the ACM Internet Measurement Conference (Montreal, QC, Canada, Oct. 2023), to appear. [DOI] [PDF] [Dataset] Details
  • Erica Stutz, Yuri Pradkin, Xiao Song and John Heidemann 2021. Visualizing Internet Measurements of Covid-19 Work-from-Home. Proceedings of the National Symposium for NSF REU Research in Data Science, Systems, and Security (REU 2021 Symposium) (Virtual Workshop, Dec. 2021), 5633–5638. [DOI] [PDF] Details
  • Xiao Song and John Heidemann 2021. Measuring the Internet during Covid-19 to Evaluate Work-from-Home. Technical Report arXiv:2102.07433v4 [cs.NI]. USC/ISI. [PDF] Details
  • Xiao Song and John Heidemann 2020. Measuring the Internet during Covid-19 to Evaluate Work-from-Home (poster). Poster at the NSF PREPARE-VO Workshop. [PDF] Details

Diurnal Block Changes

These datasets summarize changes in diurnal blocks that we detect using the algorithms described in [Song23a], section 2.

These datasets are dervied from the address accumulation datasets listed below.

name Start Date
Diurnal_Changes_a39-20200101 2020-01-01
Diurnal_Changes_a40-20200401 2020-04-01

Address Accumulation Datasets

These datasets define address changes per block over time (round by round) summarize changes in diurnal blocks that we detect using the address accumulation algorithm described in [Song23a], section 2.3.

These datasets are dervied from raw Trinocular probing datasets listed below.

See more information and format.

name Start Date
ip_accumulation_a39-20200101 2020-01-01
ip_accumulation_a40-20200401 2020-04-01

Underlying Trinocular Datasets

These are the raw Trinocular datasets used to compute address accumulation. (The are also input to the Trinocular outage detection algorithms; for outage results, please see our dataset list.)

name shortname Start Date
internet_outage_adaptive_a38w-20191001 2019q4-w 2019-10-01
internet_outage_adaptive_a39w-20200101 2020q1-w 2020-01-01
internet_outage_adaptive_a40w-20200401 2020q2-w 2020-03-01
(merge above two datasets) 2020h1-w 2020-01-01
(merge below) 2020m1-ejnw 2020-01-01
internet_outage_adaptive_a39e-20200101
internet_outage_adaptive_a39j-20200101
internet_outage_adaptive_a39n-20200101
internet_outage_adaptive_a39w-20200101
(merge above 4 and below 4) 2020h1-ejnw 2020-01-01
internet_outage_adaptive_a40e-20200401
internet_outage_adaptive_a40j-20200401
internet_outage_adaptive_a40n-20200401
internet_outage_adaptive_a40w-20200401

Getting this data

For the detailed formats of each dataset, please refer to the corresponding README file at our dataset list page.