̽��ֱ�� of Cambridge - data sharing

Is Data Justice key to Climate Justice?

plc32 — Thu, 17 Aug 2023 11:00:00 +0000

Bias in the collection of data on which Artificial Intelligence (AI) computer programmes depend can limit the usefulness of this rapidly growing tool for climate scientists predicting future scenarios and guiding global action, according to a new paper by researchers at the ̽��ֱ�� of Cambridge published in Nature’s npj |Climate Action series.

AI computer programmes used for climate science are trained to trawl through complex datasets looking for patterns and insightful information. However, missing information from certain locations on the planet, time periods or societal dynamics create “holes” in the data that can lead to unreliable climate predictions and misleading conclusions.

Primary author and Cambridge Zero Fellow Dr Ramit Debnath said that individuals with access to technology, such as scientists, teachers, professionals and businesses in the Global North are more likely to see their climate priorities and perceptions reflected in the digital information widely available for AI use.

By contrast, those without the same access to technology, such as indigenous communities in the Global South, are more likely to find their experiences, perceptions and priorities missing from those same digital sources.

“When the information on climate change is over-represented by the work of well-educated individuals at high-ranking institutions within the Global North, AI will only see climate change and climate solutions through their eyes,” Debnath said.

“Biased” AI has the potential to misrepresent climate information.

For example, it could generate ineffective weather predictions or underestimate carbon emissions from certain industries, which could then misguide governments trying to create policy and regulations aimed at mitigating or adapting to climate change.

AI-supported climate solutions which spring from biased data are in danger of harming under-represented communities, particularly those in the Global South with scant resources. These are often the same communities who also find themselves most vulnerable to the extreme weather events caused by climate change such as floods, fires, heatwaves and drought.

That is a combination which could lead to “societal tipping events”, the paper warns.

However, these “data holes” can be filled by human knowledge. ̽��ֱ��authors advocate for a human-in-the loop design to offer AI climate change programmes with a sense check on which data is used and the context in which it is used, in an effort to improve the accuracy of predictions and the usefulness of any conclusions.

̽��ֱ��authors mention popular AI chatbot model ChatGPT, which has recently taken the world by storm for its ability to communicate conversationally with human users. On ChatGPT, the AI can ask its human users follow-up questions, admit mistakes, challenge incorrect premises and reject inappropriate requests.

This ‘human-in-the-loop’ style AI allows bias to be noticed and corrected, the authors said. Users can input critical social information, such as existing infrastructure and market systems, to allow the AI to better anticipate any unintended socio-political and economic consequences of climate action.

“No data is clean or without prejudice, and this is particularly problematic for AI which relies entirely on digital information,” co-author, Cambridge Zero Director and climate scientist Professor Emily Shuckburgh said.

In highlighting the importance of globally inclusive datasets, the paper also promotes broadband internet access as a public necessity, rather than a private commodity, to engage as many users as possible in the design of AI for contemporary conversations about climate action.

̽��ֱ��paper concludes that human-guided technology remains instrumental in the development of socially responsible AI.

Less-biased AI will be critical to our understanding of how the climate is changing, and consequently in guiding realistic solutions to mitigate and adapt to the on-going climate crisis, the authors said.

Professor Shuckburgh, who also leads the UK national research funding body’s (UKRI) Centre for Doctoral Training on the Application of AI to the study of Environmental Risks (AI4ER), said that recognising the issue of data justice is the first step to better outcomes.

“Only with an active awareness of this data injustice can we begin to tackle it, and consequently, to build better and more trustworthy AI-led climate solutions,” she said.

Biased artificial intelligence needs human help to avoid harmful climate action, Cambridge researchers say.

No data is clean or without prejudice, and this is particularly problematic for AI which relies entirely on digital information

Professor Emily Shuckburgh

Photo by NOAA on Unsplash

Vehicles with weather observation equipment track a storm

̽��ֱ��text in this work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

From Mexican wave to retinal wave: why sharing data is good for science

jfp40 — Mon, 07 Apr 2014 10:40:34 +0000

Now, researchers at Cambridge, York, Newcastle and Imperial College London have developed a system allowing neurophysiologists to share raw data with each other, something they hope will generate new discoveries in the field. ̽��ֱ��results are published in the journal GigaScience.

̽��ֱ��first type of data they collected and standardised are recordings of so called ‘retinal waves’. During early development, retinal neurons generate signals that rapidly spread across from one cell to another, much like a Mexican wave in a football stadium. These patterns of activity are thought to help forge the neural connections from the eye to the brain.

To record retinal waves, scientists use multielectrode arrays (tiny electrical devices). In this research, the team took 366 recordings from 12 different studies published between 1993 and 2014, converted them all to HDF5 – a standard open source format – and published them in a web-based ‘virtual laboratory’ called CARMEN.

According to lead author Dr Stephen Eglen from the Cambridge Computational Biology Institute: “Unlike other fields such as genomics, there hasn’t been much public data sharing in neuroscience, which could be because the data are heterogeneous and hard to annotate, or because researchers are reluctant to share data with a competitor.”

But Eglen believes there is much to be gained by a more cooperative approach. “There are two main benefits to sharing,” he said. “As well as leading to other collaborations and more interesting research, it also means that other people can check what you’ve done, which leads to more robust research. And if the taxpayer funds research, then I think it’s important for those results to be publicly available.”

CARMEN was a pilot project funded by the EPSRC, and is now supported by the BBSRC.

From the way we learn, to how our memories are made and stored, the workings of our brains depend on connections forged between billions of neurons, yet much about how our nervous system develops remains a mystery.

There are two main benefits to sharing. As well as leading to other collaborations and more interesting research, it also means that other people can check what you’ve done, which leads to more robust research.

Dr Stephen Eglen

Oyvind Solstad

Eye 9

This work is licensed under a Creative Commons Licence. If you use this content on your site please link back to this page.

Yes

Licence type:

Attribution

Open and transparent

amb206 — Tue, 18 Oct 2011 13:12:56 +0000

Data“Open

̽��ֱ��Open Knowledge Foundation (OKF), a venture co-founded by a Cambridge economist and committed to the sharing of knowledge from “sonnets to statistics, genes to geodata”, is playing a pioneering role in the movement to make more information available to the public and thus empower people to become involved in decision making, whether on a local, national or international level.

Last month OKF was awarded $750,000 to develop its activities over the next three years. ̽��ֱ��award, which comes from the Omidyar Network, will help the organisation to expand its financial transparency project, sustain and build working groups on open data, and establish offshoots in a growing number of countries round the world.

OKF was set up in 2004 by Rufus Pollock, with the mission of creating a worldwide ecosystem of good quality, reliable, searchable data, giving people the tools to be able to interpret that data and increasing data-literacy. Its overall aim is to empower people, both the average citizen and policy makers worldwide, to make better decisions based on the information which is available to them.

A shining example of a new kind of hybrid not-for-profit enterprise, with both a virtual and online presence, the organisation has mushroomed into a multi-stranded enterprise – from writing the software behind national data catalogues such as the UK’s data.gov.uk to educating scientists about the importance of publishing their work in open-access journals, from increasing understanding of rights to re-use Public Domain Works to developing a set of key principles which guide decision-makers in government about how to run an open-data policy.

Its core activities are building tools and communities around an ecosystem of open-data and the guiding principle and core theme in all OKF projects is the Open Definition. According to the Open Definition: “A piece of content or data is open if anyone is free to use, reuse, and redistribute it — subject only, at most, to the requirement to attribute and share-alike.”

OKF’s next major event takes place at the end of this month. ̽��ֱ��Open Government Data Camp in Warsaw is expected to attract members of government and civil society from all over the world. Activities will include: building consensus around core open data principles and values; building community by expanding and strengthening the international open data community; sharing ideas on the future of open data and how we can do things better; and making things, from starting projects and making plans to writing code.

̽��ֱ��camp is the second of its kind and is expected to be the largest Open Data Event in the world to date with participants from over 40 different countries. This year, a special focus will be given to drafting a set of Open Government Data principles to guide decision-makers all over the globe on how to effectively implement an open-data policy. Participants will share their experiences of implementing an open-data policy in their own countries and benefit from workshops, sessions and discussions on best-practices for governments wishing to publish their data.

With its headquarters in Cambridge, OKF has a growing number of offshoots internationally. ‘Chapters’ are already established in Germany and Austria and talks are in progress about setting up chapters or partnerships in Finland, Brazil and Kenya. ̽��ֱ��flexible structure as a largely volunteer-based organisation means that people can opt in and out of its activities according to their time-constraints. Pollock remains at its helm as a director and the organisation maintains strong links with Cambridge. Pollock is affiliated to the Centre for Intellectual Property and Information Law and Peter Murray-Rust of the Department Chemistry is a key proponent of the Open-Science movement. Many key players in OKF’s Cultural Heritage projects such as have come from Cambridge’s arts and humanities faculties.

OKF works by developing standards, tools and projects, all of which are breaking new ground in the way in which they give people access to information and the opportunity to participate in building tools and forming policies. Indeed, members of the community are encouraged to set up and run their own projects and are often given infrastructure and support to do so by the Foundation.

Its tools include AnnotateIt which allows users to annotate any web page simply by incorporating two lines of JavaScript into their site or by running OKF’s bookmarklet. While some projects are targeted at the general public, and offer ways of accessing information on government spending, for example, others focus on learning. Open Shakespeare, for example is designed as a tool to allow users to explore and discover Shakespeare’s works through discovering other peoples’ perspective and commentary with tools such as AnnotateIt.

Embedded in OKF’s youthful and inclusive culture are the Principles ”, guidelines drawn up by various members, including Peter Murray-Rust in a local pub, the Panton Arms. These set out in robust terms OKF’s beliefs in the arena of open-science and include guidelines such as: “When publishing data make an explicit and robust statement of your wishes” [with regard to how you wish them to be used] and “If you want your data to be effectively used and added to by others it should be open as defined by the Open Knowledge – in particular non-commercial and other restrictive clauses should not be used.”

Pollock is currently Shuttleworth Foundation Fellow, and an Associate of the Centre for Intellectual Property and Information Law at Cambridge. He has worked extensively as a scholar and developer on the social, legal and technological issues related to the creation and sharing of knowledge. In a recent interview with the Guardian he described “vast silos of data that is not shared” and, while he acknowledged the role of technologies such as wikis, he emphasised that these were the exceptions not the rule and that a huge amount remained to be done in opening up of data. In science, for example, most research is still published in non-open journals and access is restricted to those who can afford to pay a premium for it; this is despite the fact that much of the research behind the data was publicly funded, so most readers will already have ‘paid’ for it once through their taxes.

̽��ֱ��work of OKF and others is vital in challenging our passive acceptance of the status quo, on one hand, and the obfuscation of institutions in their management of data sharing, on the other. As countries such as the UK and Germany take a lead in transparency, so others will be under pressure to follow suit. In this way OKF and similar organisations are making a valuable contribution to the sharing of data that empowers us to participate in decision-making processes, on both macro and micro levels and locally, nationally and internationally.

̽��ֱ��Omidyar Network is a philanthropic investment firm dedicated to harnessing the power of markets to create opportunities for people to improve their lives. It was established in 2004 by eBay founder Pierre Omidyar and his wife Pam. ̽��ֱ��network has since committed almost $450 million to for-profit and not-for-profit companies that foster economic advancement and individual participation.

An exciting venture dedicated to the sharing of knowledge and information, the Open Knowledge Foundation (OKF) is creating a worldwide ecosystem of searchable data and the tools to interpret that data. Founded by Cambridge economist Rufus Pollock, OKF has big ambitions in fields that range from sonnets to statistics.

Embedded in OKF’s youthful and inclusive culture are the “Pantonprinciples”, guidelines drawn up in its local pub, the Panton Arms.

Open Knowledge Foundation

Rufus Pollock

This work is licensed under a Creative Commons Licence. If you use this content on your site please link back to this page.

Yes