Project description
Searching for better search engine systems
Search engines are prime examples of information retrieval (IR) systems. Today’s IR systems use machine learning (ML) and deep learning (DL) models. Searches are based on full text or other content-based indexing. As such, IR systems rank item by relevance (semantic similarity between the user query and the information conveying items). The EU-funded METER project will extend IR evaluation measures to deal with multiple aspects. By integrating these new measures into ML algorithms, the project will develop multi-aspect IR systems. The project will also analyse IR evaluation measures to make use of differentiability in the form of properties, in order to improve the search of local minima for IR loss functions. The findings of will boost our understanding of how search engines work.
Objective
Information Retrieval (IR) deals with the automatic retrieval and ranking of information conveying items, which are relevant to a specific information need, from a large collection of items. Search engines are the most popular and well known examples of IR systems.
State-of-the-art IR systems use sophisticated Machine Learning (ML) and Deep Learning (DL) models. Those models usually minimize a loss function which is built upon an IR evaluation measure, i.e. a measure that evaluates the quality of a ranked list of items.
This project, Multi-aspect and diffErenTiable Evaluation of Rankings (METER), will tackle two open challenges for state-of-the-art IR systems. First, traditionally IR systems ranks items only by relevance, estimated as the semantic similarity between the user query and the information conveying items. However, beside relevance, understandability and trustworthines are fundamental for health search, or credibility and correctness should be considered for news search. Therefore, the first goal of METER will be to extend IR evaluation measures to deal with mutiple aspects. Then, these new evaluation measures will be integrated in ML algorithms, to develop multi-aspect IR systems.
Second, IR measures are non-continuous and non-differentiable. This represents an issue for ML algorithms, which usually exploit gradient based approaches to minize the loss function. Therefore, the second goal of METER will be to thoroughly analyze IR evaluation measures and propose differentiability like properties which will help for the search of minima of the loss function.
Therefore, METER has the potential for making both a scientific and a societal impact: 1) multi-aspect measures will be used to account for several aspect and improve the effectiveness of IR systems in different domains; 2) differentiability like properties will be exploited to improve the search of local minima for IR loss functions and better understand how this search is performed.
Fields of science
Programme(s)
Funding Scheme
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinator
1165 Kobenhavn
Denmark