The DoSSIER team from the University of Milano Bicocca participated in the Simple Text lab from Clef 2023. We focus on analysing the distribution of topics (or themes) for the content selection task and its potential effects in retrieval. We introduce Domain-knowledge through an ontology with different classification algorithms for annotating documents. This allows us to study the diversity in retrieval results and show how constraining them positively impacts the information-gathering task. We constrain results to concentrate them to specific sets of themes built using pseudo-relevance feedback.
Mendoza, O., Pasi, G. (2023). Domain Context-centered Retrieval for the Content Selection task in the Simplification of Scientific Literature. In Working Notes of the Conference and Labs of the Evaluation Forum (CLEF 2023) (pp.2979-2986). CEUR-WS.
Domain Context-centered Retrieval for the Content Selection task in the Simplification of Scientific Literature
Pasi G.
2023
Abstract
The DoSSIER team from the University of Milano Bicocca participated in the Simple Text lab from Clef 2023. We focus on analysing the distribution of topics (or themes) for the content selection task and its potential effects in retrieval. We introduce Domain-knowledge through an ontology with different classification algorithms for annotating documents. This allows us to study the diversity in retrieval results and show how constraining them positively impacts the information-gathering task. We constrain results to concentrate them to specific sets of themes built using pseudo-relevance feedback.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
