Lord777
Professional
- Messages
- 2,577
- Reaction score
- 1,563
- Points
- 113
Researchers are finding clues to understanding social change through Internet traffic.
Researchers from the Network Traffic Analysis Laboratory (ANT Lab) at the University of Southern California (USC) presented a new method for analyzing Internet traffic data that allows you to detect changes in people's behavior. The paper, published as part of the ACM Conference on Internet Measurement 2023, demonstrates how the response of IP addresses can reflect the shift to remote work and other changes in everyday life.
Since 2014, ANT Lab has been scanning the public Internet for disruptions, and since 2013, it has been actively monitoring Internet outages around the world, measuring the activity of 5 million networks every 11 minutes. This data became the basis for a new study in which scientists developed algorithms to clean up data, identify major trends, and detect changes in activity.
Using these algorithms, the researchers were able to identify the introduction of working from home due to COVID-19 in 2020, as well as other changes in people's behavior, such as national holidays and government-imposed curfews.
An example is the situation in Manila, Philippines, where there was a sharp decline in network activity at the end of March 2020, which coincided with the start of quarantine due to COVID-19. Similar changes were seen in China in January 2020, which correlated with the Wuhan lockdown and Spring Festival celebrations, while in India, changes in network activity in February and March corresponded to riots due to protests against immigration laws and the first curfew related to COVID-19.
John Heidemann, ISI's chief scientist and professor of computer science at USC Viterbi, noted that the original goal was simply to satisfy curiosity — whether it was possible to see human activity in Internet data. Now that the team has shown that this is possible, they are exploring how to use and learn from this information. In the context of COVID-19, this can help to understand which countries impose lockdowns or stay-at-home orders, when they do so, and whether people follow these orders.
This method provides a new way to understand our world, complementing other sources of publicly available information, such as news reports and surveillance of viruses in wastewater. Heidemann expressed the hope that their technique could provide a similar kind of anonymous, independent surveillance that could influence public health decisions.
The USC study provides a unique insight into how Internet data can serve as a mirror of human activity and provide valuable information for understanding social and economic processes in the world.
Researchers from the Network Traffic Analysis Laboratory (ANT Lab) at the University of Southern California (USC) presented a new method for analyzing Internet traffic data that allows you to detect changes in people's behavior. The paper, published as part of the ACM Conference on Internet Measurement 2023, demonstrates how the response of IP addresses can reflect the shift to remote work and other changes in everyday life.
Since 2014, ANT Lab has been scanning the public Internet for disruptions, and since 2013, it has been actively monitoring Internet outages around the world, measuring the activity of 5 million networks every 11 minutes. This data became the basis for a new study in which scientists developed algorithms to clean up data, identify major trends, and detect changes in activity.
Using these algorithms, the researchers were able to identify the introduction of working from home due to COVID-19 in 2020, as well as other changes in people's behavior, such as national holidays and government-imposed curfews.
An example is the situation in Manila, Philippines, where there was a sharp decline in network activity at the end of March 2020, which coincided with the start of quarantine due to COVID-19. Similar changes were seen in China in January 2020, which correlated with the Wuhan lockdown and Spring Festival celebrations, while in India, changes in network activity in February and March corresponded to riots due to protests against immigration laws and the first curfew related to COVID-19.
John Heidemann, ISI's chief scientist and professor of computer science at USC Viterbi, noted that the original goal was simply to satisfy curiosity — whether it was possible to see human activity in Internet data. Now that the team has shown that this is possible, they are exploring how to use and learn from this information. In the context of COVID-19, this can help to understand which countries impose lockdowns or stay-at-home orders, when they do so, and whether people follow these orders.
This method provides a new way to understand our world, complementing other sources of publicly available information, such as news reports and surveillance of viruses in wastewater. Heidemann expressed the hope that their technique could provide a similar kind of anonymous, independent surveillance that could influence public health decisions.
The USC study provides a unique insight into how Internet data can serve as a mirror of human activity and provide valuable information for understanding social and economic processes in the world.
