Extracting economic and social insights from Internet traffic data

By 1st July 2019 July 3rd, 2019 LINX News

Lead Data Scientist at the Office for National Statistics (ONS), gave a talk at LINX105 in May on a research project that’s exploring how different data sources can help gain insights into the data driven Internet economy.

Using data from the LINX traffic flow portal (https://portal.linx.net/) as a proxy/small window view of overall Internet traffic volume ONS have investigated the following:

  1. The potential to use traffic flow data to identify and measure Internet time-use
  2. Anomaly detection and traffic impact of large-scale events
  3. Relationship between traffic flow and various economic indicators

Philip commented:

“As part of our ongoing research in the newly formed Data Science Campus, part of the Office for National Statistics, we are currently exploring the wider theme of deriving social/economic indicators from alternative and novel data sources for potential use in new and experimental statistical outputs. One such project to emerge from this exciting research area aims to explore if it is possible to extract social and economic insights from real time internet traffic volume data. This talk summarises some insights discovered from data obtained from the LINX status monitoring portal which may be of interest to LINX members.”

The video of Philip’s talk can viewed in full below which led to much debate amongst those present at the event.

Philip Stubbings Biography

Philip Stubbings joined the ONS Data Science Campus in November 2016. He has since been applying computer vision and machine learning to image segmentation problems.

Prior to joining the campus, he had worked as a research software engineer in a number of areas including crop health monitoring, algorithmic trading, photometry, numerical control, optimisation and OR. He holds a BSc in Artificial Intelligence and MSc in Computer Science from the University of Essex.

Further Information