AI-Driven Semantic Search and Visualization to Support the Sustainable Development Goals and Agenda 2030
SDG-HUB will advance the state of the art and develop AI-based approaches to build a comprehensive knowledge repository – aiming to address the socio-ecological challenges of reaching the Sustainable Development Goals (SDGs) of the Agenda 2030 and the climate mitigation goals of the Paris Agreement in particular. The project will focus on:
- Insights into citizen and stakeholder perceptions, and into problems and value systems that determine actions of social target groups as vital input for transdisciplinary dialogues with specific societal target groups; and
- Input for monitoring and evaluation of the Agenda 2030 and Paris Agreement progress. Climate change and sustainable development communication processes will be analyzed using AI technologies like entity-centric language models that perform joint word sense disambiguation and relation extraction as well as explainable bias detection.
The AI methodology will encompass all steps from data collection to the development and training of associated neural models, and assessing their outputs. The interdisciplinary project aims to contribute measurable improvements in achieving the climate and sustainability goals in Austria.
The SDG-HUB project addresses socio-ecological challenges (“Green Aspect”) towards reaching the Sustainable Development Goals of the Agenda 2030 in general and the goals of the Paris Agreement in particular with radical innovation in the area of AI-based knowledge extraction and Explainable AI (“AI Aspect)”.
Since little has happened globally and nationally to realize the global conventions in the six years since they were signed, at the beginning of the third decade of the 21st century we are faced with the realization that both, climate and sustainability goals, can only be achieved through rapid and targeted action. Even though little has been done in Austria to overcome the challenges, there are two bottom-up initiatives supported by universities, the CCCA (Climate Change Centre Austria) and UniNEtZ (Universities and Sustainable Development Goals), which have formulated concrete proposals for solutions throughout Austria and now want to contribute to their implementation. New AI technologies will help to monitor and evaluate their implementation by political decision-makers and to advance the simultaneously planned transdisciplinary dialogue with social target groups.
Through the interdisciplinary cooperation between application and transformation-oriented climate change and sustainability research and AI, new ground is being broken on both sides and the specific research fronts are being advanced. On the climate change and sustainability-driven side, this happens through new methodological progress in the young field of the third mission resulting from insights through the SDG-HUB cooperation, i) into citizen and stakeholder perceptions, and ii) into problems and value systems that determine actions of social target groups.
On the technology-driven AI side, SDG-HUB will build on the technologies and the Web intelligence platform developed for the United Nations Environment Programme (UNEP) and advance the state of the art in several key areas: Increasing the accuracy of automated methods to extract factual and affective knowledge will require advances in the automated construction of Knowledge Graphs (KGs) that build upon joint word sense disambiguation and relation extraction algorithms. SDG-HUB will particularly focus on explainable bias detection and appropriate visual means to better understand the processing pipeline, and how errors might result from the complex interplay of several workflow components.
In detail, successfully tackling these challenges requires radical innovation and an intelligent layer that connects and enriches the fragmented pieces of information, using Explainable AI for Natural Language Understanding to support the SDGs and the Agenda 2030. The project aims to increase transparency and address the fragmented character of SDG resources by integrating (i) controlled vocabularies and open knowledge graphs, (ii) curated annotations and classifications and (iii) automated methods to annotate senses, entities, relations and to identify potential concept drift; and (iv) methods to cross-link SDG datasets and large amounts of digital content.
As an overall result, the interdisciplinary cooperation in SDG-HUB contributes to concrete improvements in achieving the climate protection and sustainability goals in Austria and thus also serves as an example beyond the national borders.
Project Start: 01 May 2022; Coordinator: webLyzard technology; Project partners: Modul Technology, Leopold-Franzens Universität Innsbruck, ZAMG – Zentralanstalt für Meteorologie und Geodynamik. The SDG-HUB project is funded by the Austrian Federal Ministry for Climate Action, Environment, Energy, Mobility and Technology (BMK) via the AI for Green Program – FFG No 892212.