Using artificial intelligence for identifying and classifying online information related to significant events

This innovation voucher project examines the feasibility of introducing AI techniques into the TEAL web application, which will search and classify news and web pages related to significant climate events, to give context for users of the app.


Established in 2015, the World Energy and Meteorology Council (WEMC) works with stakeholders around the world to promote and enhance interaction between the energy industry and the weather, climate and broader environmental sciences community.

The need to raise awareness about our changing climate has led to the development of the TEAL tool. TEAL is a newly developed and free-to-use online educational tool, working as an interactive visual front-end portal to access, display, explore and analyse selected datasets from the Climate Data Store (CDS) of the Copernicus Climate Change Service (C3S). Their data has been further developed by a team of experts to be more easily accessible to a non-technical audience.

The purpose of this feasibility study was to investigate how to search and add more contextual information to make the climate data and visualisations, TEAL offers even more informative and attractive to the general public.


The Challenge


WEMC has a proven track record of delivering such tools, with the completion of the European Climatic & Energy Mixes (ECEM) Demonstrator and more recently, the C3S Climate Education Demonstrator.

Currently, TEAL is in its first iteration and offers an attractive and accessible web interface, based around an interactive global map. With WEMC’s experienced team of climate scientists and software engineers, development of the tool has resulted in a robust platform from which to proceed, with some more bespoke software features – currently unavailable on any other tool. These are hoped to become the USP for TEAL and also the basis of more tools for different areas of the climate and energy industries.

The main goal of this study was to determine how to associate the climate data TEAL manages with real world news stories for the general public. In this manner, the casual user of TEAL can easily see how the historical data TEAL offers is associated with the related climate events that had produced various impacts on the environment, social and public life reported in news articles and possibly to aid in research findings.

The objectives of this project were to:

  • Investigate, evaluate or devise the necessary AI technology or research that can search for and classify information on the internet (e.g news, web pages) related to significant climate events
  • To investigate the feasibility of integrating the AI technologies into TEAL
  • To produce a roadmap for implementation.

At the moment, TEAL and other similar tools are only able to present the chosen climate data and events (storms, flooding, heatwaves, etc.) in the usual abstract, scientific, and mostly numerical way. This approach lacks more meaningful real-world contextual information such as the influence on the environment, society, and ordinary people. If the information was simplified to emphasize humanity’s relationship to the climate and presented in a more user friendly manner, it would certainly add more value and attract a wider audience.

With such AI functionality, TEAL will bring relevant, real-world context to the data in a timely fashion, attract more users and hence increase the awareness of climate change and its impact among the public.


The Approach


Dr. Wenjia Wang, an Associate Professor of Artificial Intelligence at the School of Computing Sciences, the University of East Anglia, has been leading his team in researching AI and machine learning for many years, with a focus on developing ensemble methods and AI systems for various applications.

For this project, he designed 4 work packages:

  • Understanding the climate data and familiarisation with TEAL
  •  Identifying major climate events in the data
  • Investigating and evaluating AI technologies for search and classification of text documents
  • Preliminary design of the IA framework/components.

After completing the first three tasks, Dr. Wang and his Senior Research Associate, Doug Fraser, designed a framework with essential AI functionalities to enhance user experience (UX) by providing more contextual information for the climate event of user’s interest.

The framework contains 4 main components:

  • User Interface – it should allow a user to choose the climate event of interest (CEOI) from the TEAL to start their search
  • Relevant information collection – with the chosen event, it searches the internet and retrieves the relevant information (news reports, articles, dates etc)
  • Data processing and feature extraction – it processes the retrieved unstructured textual data with natural language processing techniques and transforms them into a structured representation
  • Classification – the retrieved and transformed information is classified into some predefined categories ready for presenting them back to the user.

The outcomes for this project was the delivery of a prototype software/script of the developed algorithm that demonstrates the concept and a full technical report.


The Outcome


A full report was written which outlined proposed enhancements to the existing TEAL system, the different components needed and the feasibility of the natural Language Processing (NLP) techniques used to categorize and extract information from articles.

A prototype software system ‘Spiderman’ was developed.

The project demonstrated that it is feasible to apply suitable AI technology/packages and other necessary methods or tools to build a relatively independent software module of the proposed AI system to plug into the existing web system TEAL.

Luke Sanger, Data Engineer, World Energy & Meteorology Council said:

“The case study provided an excellent groundwork in which to further our development in the field of artificial intelligence and climate science. We are looking forward to continue working with UEA in the future!”

WEMC benefitted from the project in 2 ways. Firstly, receiving a working prototype piece of code that can be adapted for future needs. Secondly, UEA provided expert advice on other development queries related to the project, which have been adopted by the team going forward.

Dr Wenjia Wang, who led the project, said:

“Thanks to EIRA, with the support from their innovation voucher, we have an opportunity to work with WEMC to conduct a feasibility study in developing an AI technology to help enhance their user’s experience through providing more contextual information related to a climate event of their interest. In this short study, we have developed a framework that has been demonstrated to achieve all the objectives. This work has laid a solid foundation for further research and development that should integrate this framework into WEMC’s TEAL and hence deliver more enjoyable and informative experience to their users.”


Next Steps


The outcomes from this project will feed directly into implementing the AI technology into the TEAL tool with the aim of becoming one of the USPs of TEAL and enabling WEMC to approach more commercial contracts based on the TEAL platform, in addition to strengthening their reputation amongst the climate science community and energy industry.