Assessing the social impacts of extreme weather events using social media
Lead Research Organisation:
University of Exeter
Department Name: Engineering Computer Science and Maths
Abstract
Natural hazards cause major disruption to the UK economy, but their impacts are hard to forecast and observe. This project will develop methods for using social media data to accurately map natural hazards and their impacts. "Social sensing" can be defined as observation of real-world events using unsolicited content from digital communications (e.g. mobile phone call records, social media, web searches, and other online data). The challenge of social sensing is to extract high-quality observations from large numbers of unstructured, patchy, and possibly inaccurate user utterances. If this can be achieved then there are significant opportunities to use this data in areas where observations are not currently available.
One key application area is assessing the impact of natural hazards, where forecasts are routinely produced based on meteorological models, but for which little impact observation data is available for model validation.
Previous work by the investigators has shown that social media can be used successfully as a source of data with which to detect and locate wildfires, floods and extreme rainfall events. However, the methods are at an early stage of development. This project will establish robust methods for two key aspects of the social sensing pipeline for natural hazards:
(1) Content classification. The first part of the project will use machine learning to create text-based classifiers that can automatically categorise social media posts based on their content.
(2) Location inference. Identifying the geographical origin of social media content is essential for hazard impact evaluations, but only a small fraction of social media posts include accurate geotags.
The second part of the project will develop effective machine learning methods for inferring the locations of un-geotagged social media posts.
One key application area is assessing the impact of natural hazards, where forecasts are routinely produced based on meteorological models, but for which little impact observation data is available for model validation.
Previous work by the investigators has shown that social media can be used successfully as a source of data with which to detect and locate wildfires, floods and extreme rainfall events. However, the methods are at an early stage of development. This project will establish robust methods for two key aspects of the social sensing pipeline for natural hazards:
(1) Content classification. The first part of the project will use machine learning to create text-based classifiers that can automatically categorise social media posts based on their content.
(2) Location inference. Identifying the geographical origin of social media content is essential for hazard impact evaluations, but only a small fraction of social media posts include accurate geotags.
The second part of the project will develop effective machine learning methods for inferring the locations of un-geotagged social media posts.
Organisations
People |
ORCID iD |
Hywel Williams (Primary Supervisor) | |
Michelle Spruce (Student) |
Publications
Spruce M
(2020)
Using social media to measure impacts of named storm events in the United Kingdom and Ireland
in Meteorological Applications
Spruce M
(2021)
Social sensing of high-impact rainfall events worldwide: a benchmark comparison against manually curated impact observations
in Natural Hazards and Earth System Sciences
Young J
(2021)
Social Sensing of Heatwaves
in Sensors
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/N509656/1 | 01/10/2016 | 30/09/2021 | |||
1939288 | Studentship | EP/N509656/1 | 01/10/2017 | 31/03/2022 | Michelle Spruce |
EP/R513210/1 | 01/10/2018 | 30/09/2023 | |||
1939288 | Studentship | EP/R513210/1 | 01/10/2017 | 31/03/2022 | Michelle Spruce |
Description | Despite increasing use of impact-based weather warnings by meteorological agencies, the social impacts of extreme weather events lie beyond the reach of conventional meteorological observations and remain difficult to quantify. As a result of the work funded through this award, the use of social media data (in particular Twitter) for detecting and locating specific extreme weather events has proved to be successful. Social media data is an unstructured data source which is difficult to filter for specific events. By the application of 'social sensing', the systematic analysis of unsolicited social media data to observe real-world events, it has been possible to apply a bespoke filtering process on unfiltered social media data to detect and locate extreme weather events (named storms in the UK, heavy rainfall events globally). The social media data obtained has then been used to begin to better understand the social impacts of these events. This has provided a novel methodology for identifying social media data which can be used to assess the impacts of extreme weather events. |
Exploitation Route | Further development of the outcomes of this work could lead to improved understanding of the social impacts of extreme weather events and weather forecast impact model validation. This could be in the form of a retrospective analysis of social media data to aid in the evaluation of weather impact models, or in the form of a live tool which allows a user to monitor social media data relating to extreme weather events and/or their impacts in real-time. |
Sectors | Communities and Social Services/Policy,Digital/Communication/Information Technologies (including Software),Other |
Company Name | SOCIAL SENSING LTD |
Description | Social Sensing Ltd provides access to an online application which utilises social media data (from Twitter) to monitor activity in the UK and Ireland relating to reports of flooding, winds/storms and snow. |
Year Established | 2019 |
Impact | Our Social Sensing application is in operational use in the Met Office, Environment Agency and Natural Resources Wales. |
Website | https://socialsensing.com |
Description | Interview for BBC radio show |
Form Of Engagement Activity | A press release, press conference or response to a media enquiry/interview |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Media (as a channel to the public) |
Results and Impact | Michelle was invited to be a guest on the BBC Radio Paul Hudson Weather Show to talk about her research on using social media to identify impacts from named storm events in the UK. Her interview with Paul was broadcast on the show in December 2018. |
Year(s) Of Engagement Activity | 2018 |
Description | Invited speaker at public conference |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | Michelle was an invited speaker to the Royal Meterological Society's public 'Weatherlive' conference for it's members in November 2018. Michelle spoke about her research utilising Twitter data to detect and locate extreme weather events both in the UK and globally. |
Year(s) Of Engagement Activity | 2018 |
URL | https://www.rmets.org/weatherlive |