Wind speed is something that is hard to qualitatively describe from person to person. One person's playful breeze is another's biting annoyance. For a very long time this was a problem in the maritime domain so a couple of centuries ago Francis Beaufort came up with the Beaufort Scale. This scale ties wind speed to objective observations. It was first applied to sea state, but it was later extended for observations on land.
There is a research field growing up around the idea of "social sensing", where social media platforms, particularly Twitter, can be used for real-time detection and tracking of natural events, such as earthquakes, forest fires, air quality, etc. A group of researchers from the University of Exeter have established a social Beaufort scale using Twitter. They looked at 110k weather-related tweets in the UK spanning two years to see if they could detect wind-related effects and estimate the wind magnitudes by looking at the language and emojis used, similar to what is done with the Beaufort scale (well, except for the emojis). They found that a simple text classifier can be used to detect high-wind events fairly accurately and the severity of these events can be inferred by considering the tweet volume.
Journal Reference:
Iain S. Weaver, Hywel T. P. Williams, Rudy Arthur. A social Beaufort scale to detect high winds using language in social media posts [open], Scientific Reports (DOI: 10.1038/s41598-021-82808-x)
ABSTRACT:
People often talk about the weather on social media, using different vocabulary to describe different conditions. Here we combine a large collection of wind-related Twitter posts (tweets) and UK Met Office wind speed observations to explore the relationship between tweet volume, tweet language and wind speeds in the UK. We find that wind speeds are experienced subjectively relative to the local baseline, so that the same absolute wind speed is reported as stronger or weaker depending on the typical weather conditions in the local area. Different linguistic tokens (words and emojis) are associated with different wind speeds. These associations can be used to create a simple text classifier to detect 'high-wind' tweets with reasonable accuracy; this can be used to detect high winds in a locality using only a single tweet. We also construct a 'social Beaufort scale' to infer wind speeds based only on the language used in tweets. Together with the classifier, this demonstrates that language alone is indicative of weather conditions, independent of tweet volume. However, the number of high-wind tweets shows a strong temporal correlation with local wind speeds, increasing the ability of a combined language-plus-volume system to successfully detect high winds. Our findings complement previous work in social sensing of weather hazards that has focused on the relationship between tweet volume and severity. These results show that impacts of wind and storms are found in how people communicate and use language, a novel dimension in understanding the social impacts of extreme weather.
(Score: 0) by Anonymous Coward on Monday February 15 2021, @10:00AM (1 child)
Rainfort #1 Scottish Mist
Rainfort #2 Friends fall out arguing "Is this rain?"
Rainfort #3 Light Drizzle
Rainfort #4 Heavy Drizzle
Rainfort #5 Light Rain
Rainfort #6 Heavy Rain
Rainfort #7 Cats'n'dogs
Rainfort #8 Stair Rods
Rainfort #9 Small Cornish villages washed away
Rainfort #10 Large Cornish towns washed away
Rainfort #11 Paging Noah
Rainfort #12 Hosepipe ban cancelled
(Score: 0, Troll) by Ethanol-fueled on Monday February 15 2021, @08:39PM
The Chimpout scale, too:
Catagory 1: Knockout game
Catagory 2: Flash mobs
Catagory 3: Riots
Catagory 4: South Africa
Catagory 5: Planet of the Apes
(Score: 2) by FatPhil on Monday February 15 2021, @10:13AM (2 children)
Erm, nope, I've always thought that people talking more about something than typical was an indication that that something is more relevant than typical. That's my prior. This study leaves my posterior unchanged from my prior. It contains no new bits.
Great minds discuss ideas; average minds discuss events; small minds discuss people; the smallest discuss themselves
(Score: 1, Insightful) by Anonymous Coward on Monday February 15 2021, @10:30AM
Twitter is 50% social engineering bots. They are determining the wind chill now.
(Score: 2) by driverless on Monday February 15 2021, @12:04PM
This is an evaluation of wind from social media posts, the whole point is to detect changes in posteriors. For example "aww geeze Paul, couldn't you have done that outside" indicates a strong presence of wind, "look at this cute cat video" indicates little to no wind, and moderate posterior tweets indicates some wind.
(Score: 0) by Anonymous Coward on Monday February 15 2021, @12:03PM
So weather is now to follow news, climate and social correctness (aka justice, ska norms) down the gurgler.
why not treat temps the same way, for some people 25C. is as good as 15C if they are more used to daytime temps nearer 30.
that way when warming gets us to that +2 degrees we justneed to vote for "everything is fine" on the social sensometers.
(Score: 1, Touché) by Anonymous Coward on Monday February 15 2021, @12:20PM
Meteorology departments need more funding. If the government can cover London in cameras, they can cover the country with weather sensors. Or hire someone to make an official app for reporting weather conditions.
(Score: 0) by Anonymous Coward on Monday February 15 2021, @12:44PM
Congrats to whoever came up with:
from the The-wind-tweets-Mary dept.
Just took a little side trip to visit Jimi (on old video).
(Score: 2) by Runaway1956 on Monday February 15 2021, @01:54PM (1 child)
I would have thought it took five years or more to find 110k tweets that weren't about fog and overcast.
“I have become friends with many school shooters” - Tampon Tim Walz
(Score: 1, Funny) by Anonymous Coward on Monday February 15 2021, @03:40PM
Go count how many words for rain there are in the English language.
(Score: 2) by looorg on Monday February 15 2021, @03:41PM
So is this like weather forecasting outsourcing? 110k twitts over a two year period is not exactly a lot of data to base your model on, they might not be evenly spaced out in time and geography. Also it becomes highly subjective as people seem to feel different about the weather -- what is cold, windy and what not. They might as well just look for images online and try and identify things blowing in the wind and from that judge wind-speed. It would probably be better.
The weather man will no longer tell us that weather but instead just stand there and read tweets about how people felt the weather was today ... and you won't get to know tomorrows weather until tomorrow evening or is he perhaps just going to read them out loud live on some kind of 24-7 tweet-weather-channel?
(Score: 2) by nostyle on Monday February 15 2021, @10:34PM
...You don't need a weather man
To know which way the wind blows....
---
Bob Dylan, "Subterranean Homesick Blues"