This document demonstrates the work of the LGTM Team of the University of Padua’s participation in CLEF 2022 Touchè Task 1, creating a system to retrieve arguments for controversial questions. This system aims to provide a solution to assist users who search for arguments to be used in conversations and retrieves a pair of sentences from a collection of arguments. In the various runs, we used a custom parser for parsing the document collection, a custom indexer, the BM25 Similarity, the DataMuse API for query expansion and boosting in Searcher. We also tried to improve the system by using different stoplists, stemmers and apply re-ranking based on sentiment analysis and readability of the document.
SEUPD@CLEF: Team Lgtm on Argument Retrieval for Controversial Questions
Ferro N.
2022
Abstract
This document demonstrates the work of the LGTM Team of the University of Padua’s participation in CLEF 2022 Touchè Task 1, creating a system to retrieve arguments for controversial questions. This system aims to provide a solution to assist users who search for arguments to be used in conversations and retrieves a pair of sentences from a collection of arguments. In the various runs, we used a custom parser for parsing the document collection, a custom indexer, the BM25 Similarity, the DataMuse API for query expansion and boosting in Searcher. We also tried to improve the system by using different stoplists, stemmers and apply re-ranking based on sentiment analysis and readability of the document.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
