urbanteacher.bsky.social - Profile | ThreadSky | a Reddit-style client for Bluesky

Pie and donut charts get a bad rep, but they work well if used for the right data and tasks. Read about what the science has to say about them in our new blog post: https://buff.ly/3DURbnS

submitted 43 days ago • 0 comments

👀

submitted 11 days ago • 0 comments

Scientists have developed a solar-powered device that can pull pollution from the air and convert it directly into fuel for cars & planes. The new reactor takes its inspiration from photosynthesis, requiring no cables in order to turn CO2 into syngas. www.the-independent.com/tech/cambrid...

submitted 12 days ago • 0 comments

A tool that lets you design metro maps Metro Designer - Tennessine tennessine.co.uk/metro/ #maps #edusky

submitted 13 days ago • 0 comments

As US and UK refuse to sign the Paris AI Action Summit statement, other countries commit to developing ‘open, inclusive, ethical’ AI

submitted 16 days ago • 1 comment

Note: Propaganda or Misinformation is not always loud or obvious. Often, it’s wrapped in half-truths, emotional hooks, or shared by people you follow. Are we really that good at recognising manipulation?

submitted 17 days ago • 0 comments

Note: Algorithms don’t care about truth—they care about engagement.

submitted 17 days ago • 0 comments

Note: Algorithms love extremes—outrage, fear, blind loyalty.

submitted 17 days ago • 0 comments

A recent study found that even short-term exposure to air pollution—like PM from something as common as candle smoke—can impair cognitive focus. "How much pollution from busy roads seeps into our classrooms? How does it impact learning?" #edusky www.theguardian.com/environment/...

submitted 19 days ago • 1 comment

Designing public compute for people and society www.adalovelaceinstitute.org/report/compu...

submitted 20 days ago • 0 comments

‘Godfather of AI’ predicts it will take over the world youtu.be/vxkBE23zDmQ?...

submitted 24 days ago • 0 comments

New tutorial paper published today with PhD student @danaroemling.bsky.social and @bodowinter.bsky.social Visualizing map data for linguistics using ggplot2: A tutorial with examples from dialectology and typology doi.org/10.1017/jlg.... I hope people find it useful!

submitted 28 days ago • 4 comments

📼 New EAS tutorial video from @codewithbeto.dev for learning how to: • work with environment variables, • build and update specific app variants • manage channels for streamlined development and deployment youtu.be/uKGx3gRrhx0

submitted 28 days ago • 1 comment

"We need to see Algorithms as a power structure, designed to shape our thinking. Algorithms are not just technical tools; they are power structures intentionally designed to shape our thinking, influence our behavior, and reinforce societal norms." www.urbanteacher.co.uk/the-dark-mir...

submitted 26 days ago • 0 comments

London Underground: Which Tube lines are the most polluted. #AirPollution Full Article www.ft.com/content/6f38...

submitted 27 days ago • 0 comments

Hot off the press! 📣📣In this tutorial we illustrate available multiple imputation approaches for handling longitudinal data including when they are clustered within higher level clusters. A reproducible example with R and Stata code provided! #OpenAccess onlinelibrary.wiley.com/doi/10.1002/...

submitted 31 days ago • 5 comments

Five ways to effectively use animation in data visualization #datascience #computerscience #bigdata observablehq.com/blog/effecti...

submitted 29 days ago • 0 comments

Air Pollution: not just an outdoors problem link.medium.com/meqC42MGHyb

submitted 30 days ago • 0 comments

📈📉We've just published a nifty new #ONS tool that allows you to see how different factors affect population projections! Adjust life expectancy, fertility rate and net migration to see how they affect estimates... www.ons.gov.uk/peoplepopula...

submitted 30 days ago • 2 comments

Ok I've finally managed to publish this easy piece on #probability, going through the Bernoulli - give me any feedback, tell me how it is, let me know if you find any mistakes! martinapugliese.github.io/excursus/pro...

submitted 31 days ago • 3 comments

submitted 30 days ago • 1 comment

What a day! DeepSeek just dropped another AI model, Janus Pro 7B This time an open source text-to-image model huggingface.co/deepseek-ai/...

submitted 31 days ago • 0 comments

The reveal of a new artificial intelligence assistant, DeepSeek, looks poised to wipe almost a trillion pounds in value off some of the world’s most expensive technology companies.

submitted 31 days ago • 52 comments

DeepSeek ‘punctures’ tech spending plans, and what analysts are saying

submitted 31 days ago • 4 comments

An international team comprised of 23 researchers has published a review article on the future of neuromorphic computing that examines the state of neuromorphic technology and presents a strategy for building large-scale neuromorphic systems. www.genengnews.com/topics/artif...

submitted 31 days ago • 0 comments

Can a neural network learn to recognize doodling? Check this website to help teach it by adding your drawings to the world’s largest doodling data set, shared publicly to help with machine learning research. quickdraw.withgoogle.com #edusky #computerscience

submitted 32 days ago • 1 comment

Ever wanted to see what was happening in Large Language Models (LLM) like ChatGPT? This 3D LLM visualisation tool shows all the steps to run a single token inference. Website: bbycroft.net/llm #edusky #ai #llm

submitted 32 days ago • 0 comments

💡Imagine an AI that runs on pure sunlight: A solar-powered local LLM that learns, thinks, and computes using renewable energy!

submitted 32 days ago • 0 comments

Deepseek R1 for Everyone 💡 trite-song-d6a.notion.site/Deepseek-R1-...

submitted 32 days ago • 0 comments

Spatial Statistics for Data Science: Theory and Practice with R by Paula Moraga #RStats https://bigbookofr.com/chapters/geospatial.html#spatial-statistics-for-data-science-theory-and-practice-with-r

submitted 38 days ago • 0 comments

"There's an old saying: if it looks like an AI agent, talks like an AI agent... it's probably a web scraper. 🤖🕸️ #AI #WebScraping"

submitted 40 days ago • 0 comments

If you hate statistics like I do, then you'll love my free lectures. Putting science before statistics, 20 lectures from basics of inference & causal modeling to multilevel models & dynamic state space models. It's all free, made with love and sympathy. 🧪 #stats www.youtube.com/playlist?lis...

submitted 161 days ago • 23 comments

We took 100+ hours of footage of Lexington Ave in Manhattan, capturing some 75,000 vehicles before and after congestion pricing started, and found the drop in vehicles appears to be coming entirely from fewer personal cars. Also no change in avg car value. Congestion pricing is working (for now).

submitted 41 days ago • 53 comments

Are you interested in the new story map feature in {mapgl}, but all of your #rstats mapping already uses Leaflet? Good news: the package's story map feature supports Leaflet as well! Check out this video - it's an example from the R Leaflet docs translated to a story map.

submitted 41 days ago • 2 comments

Apple halts Apple Intelligence news summarization service after multiple hallucinations www-bbc-co-uk.cdn.ampproject.org/c/s/www.bbc....

submitted 41 days ago • 0 comments

Brussels orders X to hand over documents on algorithm https://www.ft.com/content/a6dc562c-4fa0-4ec6-9f3a-ad3be594bc7c

submitted 41 days ago • 8 comments

Gamification Model of Learning 💡 #EduSky

submitted 41 days ago • 0 comments

Learn difficult concepts with the ADEPT Method #EduSky

submitted 41 days ago • 0 comments

Note: Did you know the word 'broadcast' originates from farmers scattering seeds across their fields? Just like those seeds, when media / algorithm broadcasts information, each word is a seed planted in the public mind - carefully chosen to sow either harmony or discord.

submitted 43 days ago • 0 comments

Houston we have a problem. This creates black hole for real applications and forces recruiters to make terrible decisions.

submitted 48 days ago • 0 comments

🚨 New blog post! 🚨 If you want to learn about: 🎨 How to define colours in R 📊 Choosing palettes for plots 🛠️ Generating your own colours Read this ➡️ nrennie.rbind.io/blog/colours... #DataViz #RStats #ggplot2 #RLadies

submitted 48 days ago • 5 comments

Note: Misinformation is designed to provoke an emotional response, sparking anger or outrage. Users impulsively pass it on without checking its validity or intentionally to clout chase.

submitted 51 days ago • 0 comments

Lovely example of how weird real-world data is. I put all 55 million points in Overture maps into a Nomic Atlas map. Lots of points spuriously appearing in the ocean. But the 1° square around 'null island' (0° lat, 0° long) is suspiciously clear -- someone has cleared out bad data just there.

submitted 52 days ago • 9 comments

"This moment reminds me of the calculator debate of the 1970s. When pocket calculators became widely available, critics feared they would erode basic arithmetic skills. Yet today, while calculators are ubiquitous, mathematical skills remains essential" #edusky www.urbanteacher.co.uk/knowledge-do...

submitted 68 days ago • 0 comments

A new study finds ants best humans at tests of collective intelligence. Learn more: scim.ag/4h2K0ID

submitted 55 days ago • 27 comments

Some of my thoughts on OpenAI's o3 and the ARC-AGI benchmark aiguide.substack.com/p/did-openai...

submitted 66 days ago • 17 comments

Phil Schmid’s tutorial on fine tuning ModernBERT for classification (hope this is easy with prodigy/spacy) www.philschmid.de/fine-tune-mo...

submitted 63 days ago • 2 comments

Ever wanted to waste infinity hours doing tile puzzles of every place in the world? Well, now you can with #AnyMapPuzzle You can play for free, forever (literally), at bothness.github.io/anymap

submitted 64 days ago • 8 comments