📣 Help Shape the Future of UKRI's Gateway to Research (GtR)

We're improving UKRI's Gateway to Research and are seeking your input! If you would be interested in being interviewed about the improvements we're making and to have your say about how we can make GtR more user-friendly, impactful, and effective for the Research and Innovation community, please email gateway@ukri.org.

OMINO - Overcoming Multilevel INformation Overload

Lead Research Organisation: University of Wolverhampton
Department Name: Faculty of Science & Engineering

Abstract

In today's world, access to information thought of as the resolution of uncertainty; is often considered a benefit or even an indisputable human right. There is, however, the "dark side" of information: the abundance of data beyond one's capacity to process them leads to so-called information overload (IOL). This notion had troubled mankind long before even the print was invented and examined from different points of view, ranging from neuroscience to journalism. IOL is, however, usually considered at the individual level by examining a single factor or a specific level that eventually leads to switching off an active individual. The influence of IOL appearing simultaneously at different levels, i.e., a multilevel information overload is unknown, though. These observations lead to setting the main aim of the OMINO - Overcoming Multilevel INformation Overload project in a form of the following objectives: (1) create and apply means to measure multilevel IOL in different systems as well as methods to model IOL and counter-measures to mitigate this phenomenon, (2) training and knowledge exchange on IOL between partners in different domains using expertise from universities in U.S., Singapore and Japan, (3) intersectoral knowledge transfer between academia and the media industry (Slovenian and Austrian Press Agencies) by exposing researchers to real-life problems and giving business access to innovative methods and tools for information analysis. One of the most important aspects of the undertaken research area is its interdisciplinary nature, requiring joint work of experts in different fields and topics, i.e., social sciences, neuroscience, journalism, computing, data mining and complexity science. OMINO will accelerate the individual careers of involved researchers, especially early-stage ones and increase their employability. The project will lay foundations for long-term collaboration by strengthening existing links between partners and creating new ones.
 
Description The work so far focused mainly on the implementation and integration of AI methods to the GESIS search engine, which is a search engine for social scientists. AI-based summarisation and LLM-based chatbots like ChatGPT play an increasing role to find relevant information. Our prototype, which will be evaluated in the second half of the project, integrates these new technologies. Our evaluation will focus on the question whether this helps reducing information overload. Future work will include the integration of retrieval-augmented generation and conversational agents.
Exploitation Route AI-based summarisation and conversational agents are a new mode to access information, which is able to tackle information overload (by providing a quick overview of search results) but also provide immediate access to original sources (the search engine functionality). This way, professional users can access the original sources whilst benefitting from AI-base guidance through the search results.
Sectors Communities and Social Services/Policy

Digital/Communication/Information Technologies (including Software)

Education

URL https://searchtest.gesis.org/?source=%7B%22query%22%3A%7B%22bool%22%3A%7B%22must%22%3A%5B%7B%22query_string%22%3A%7B%22query%22%3A%22migration%22%2C%22default_operator%22%3A%22AND%22%7D%7D%5D%2C%22filter%22%3A%5B%7B%22term%22%3A%7B%22type%22%3A%22research_data%22%7D%7D%5D%7D%7D%7D&lang=en
 
Description House of Lords Communications and Digital Select Committee inquiry (FON0031)
Geographic Reach National 
Policy Influence Type Contribution to a national consultation/review
URL https://committees.parliament.uk/writtenevidence/128380/html/
 
Title GESIS Advanced Search Engine 
Description We enriched the GESIS search engine (https://search.gesis.org/) with Large Language Models, for example to provide summaries of search results. 
Type Of Material Improvements to research infrastructure 
Year Produced 2024 
Provided To Others? No  
Impact The tool will help scientists to find relevant literature and data more effectively, thus reducing information overload. 
 
Description GESIS Collaboration 
Organisation Leibniz Association
Department Leibniz Institute for the Social Sciences
Country Germany 
Sector Academic/University 
PI Contribution We established a fruitful collaboration with GESIS, the Leibniz Institute for the Social Sciences (http://gesis.org/). We created a service based on Large Language Models for the GESIS Search engine. Furthermore, we established a joint seminar series for talks around information retrieval and AI. The collaboration also led to the continuation of the successful BIR workshop series in 2023 and 2024 (https://sites.google.com/view/bir-ws/bir-2024) and the preparation of several research articles.
Collaborator Contribution Provision of infrastructure and expertise. Handling of a seminar. C
Impact Two Bibliometric-enhanced Information Retrieval workshops at ECIR (2023 and 2024, see https://sites.google.com/view/bir-ws/bir-2024). Design and implementation of search services based on Large Language Models.
Start Year 2023
 
Description Modul University, Vienna, Austria 
Organisation Modul University Vienna
Country Austria 
Sector Academic/University 
PI Contribution We conducted a seminar about our work in Vienna and discussed further activities regarding issues around bias and privacy.
Collaborator Contribution Hosting our researchers, conducting a seminar, providing expertise.
Impact Knowledge exchange
Start Year 2023
 
Title Extended GESIS Search 
Description We extended the GESIS Search Engine for social scientists with large language models, e.g., to summarise search results. This is implemented as Web-based micro services. 
Type Of Technology Webtool/Application 
Year Produced 2023 
Impact We hope our tool will help scientists to find relevant data and documents more effectively.