1. General

Category

Please describe Other

We are currently tackling SDG11, but in the future, we will tackle other goals as well.

Category

Other

Please describe Other

We work with different stakeholders. For example, for SDG11, we addressed governments andprivate companies that work with city data, such as improving cycling in cities, tackling environmental issues, and enhancing mobility.

2. Project Details

Company or Institution

World Data League

Project

World Data League (WDL)

General description of the AI solution

We are a global Data and AI competition that provides opportunities for open innovation with participants worldwide. We work directly with entities (NGOs, companies, and governmental organizations) that have social impact challenges that can be solved throughdatat. All of our challenges are under the UN SDGs. At the end of the competition, we provide several proofs-of-concept, available  under the MIT licence, that anyone can use for further development. We also write a report summarising all the outcomes published under the Creative Commons licence.With these actions, we hope to increase the impact of the competition. The competition is also an excellent opportunity for organisations to test the capacities of the datasets.

Website

https://www.worlddataleague.com

Organisation

World Data League

3. Aspects

Excellence and Scientific Quality: Please detail the improvements made by the nominee or the nominees’ team or yourself if your applying for the award, and why they have been a success.

The participants have a time limit to develop the AI solutions based on data to the partner entities’ challenges. The solutions are composed of a needs analysis, a technical proposal with a working model, and an impact analysis. The type of AI that is implemented varies a lot depending on the project, for example,computer vision, churn prediction, and spatio-temporal time series forecasting have already been used.

The research work is detailed in Jupyter Notebooks, with several textual and image descriptions, along with charts and other visualizations. All conclusions, future improvements, and social impact measurement are also described in this format. Besides this, all research work is also accompanied by a 1-minute video describing the main findings and an evaluation from several jury members, following stringent evaluation criteria. We encourage teams to submit their work to scientific conferences after the competition. In 2021, a paper was accepted at a ECML-PKDD workshop.

Scaling of impact to SDGs: Please detail how many citizens/communities and/or researchers/businesses this has had or can have a positive impact on, including particular groups where applicable and to what extent.

In the first edition, we worked with four cities – or labs that work with those cities -: Medellin, Lisbon, Porto, and Cascais. If the solutions are implemented, that is a total impacted population of 3.5M.

To measure the tangible impact of the solutions proposed by the participants, we follow up with the entities that provided the   to measure the quality and viability. We will also discuss if there is a need to resubmit the challenge next year with improvements.

In order to reach a wider audience and impact, all of the solutions will be  open-source under the MIT licence, and an insights report on this edition’s success and failures was published and disseminated through different channels.

Scaling of AI solution: Please detail what proof of concept or implementations can you show now in terms of its efficacy and how the solution can be scaled to provide a global impact ad how realistic that scaling is.

One of the evaluation criteria of the solutions proposed by the participants is precisely the solution’s scalability and its possible application in other contexts. This means that every solution was evaluated and rated on a scale of 0 to 3 on this criteria, by at least four senior jury members. In a year, we will assess the reach of the proposed solutions.

The last edition of WDL had 186 participants from 32 different countries. Next year we anticipate increasing these numbers by partnering up with other data competition organisers that propose challenges regarding the SDGs, and more AI tech communities and universities.

By having open-source as one of our core values, we envision that this will lead to significant growth of our ecosystem and outreach, both on the technical/AI and  social impact aspects. We want institutions to look at us as a competition where their problems get solved through AI.

Ethical aspect: Please detail the way the solution addresses any of the main ethical aspects, including trustworthiness, bias, gender issues, etc.

In the case of the World Data League, there are two-folds of ethical implications that we considered:

1. Organisational – how we, as organisers, guarantee the ethical aspects in the competitions. For this, we strived to have diverse participants from different nationalities. We implemented a quota system that allowed us to have up to a limit of participants of a certain nationality, and we also contacted several technical communities which represented gender-based communities (e.g., women in data science) and geographical communities (e.g., Data Science Summit Africa). We also contacted several topic entities worldwide (e.g., Medellin, Torino, and Lisbon). The competition challenges also incorporate population diversity (e.g., public transportation for the elderly).
2. AI solution – how participants’ proposed solutions also consider the ethical aspects. For this, we evaluate if the participant explores the fairness and bias of the model or reports any ethical considerations. This means that every solution was evaluated and rated on a scale from 0 to 3 on this criteria. Teams were expected to provide solutions to make their models more transparent and explainable.