Conversational Search with Random Walks over Entity Graphs
Open Access
- 9 August 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 77-85
Abstract
The entities that emerge during a conversation can be used to model topics, but not all entities are equally useful for this task. Modeling the conversation with entity graphs and predicting each entity's centrality in the conversation provides additional information that improves the retrieval of answer passages for the current question. Experiments show that using random walks to estimate entity centrality on conversation entity graphs improves top precision answer passage ranking over competitive transformer-based baselines.Keywords
Funding Information
- CMU|Portugal (LISBOA-01-0247-FEDER-045920, CMUP-ERI/TIC/0046/2014)
- Fundação para a Ciência e Tecnologia (UIDP/04516/2020)
This publication has 19 references indexed in Scilit:
- Multilingual Autoregressive Entity LinkingTransactions of the Association for Computational Linguistics, 2022
- Contextualizing Trending Entities in News Stories2021
- A Hybrid Retrieval-Generation Neural Conversation Model2019
- BERT with History Answer Embedding for Conversational Question Answering2019
- Towards Better Text Understanding and Retrieval through Kernel Entity Salience Modeling2018
- Benchmark for Complex Answer Retrieval2017
- Fast and Accurate Annotation of Short Texts with Wikipedia PagesIEEE Software, 2012
- DBpedia spotlight2011
- On Variations of Power Iteration2005
- Deeper Inside PageRankInternet Mathematics, 2004