Shadowfax: Harnessing Textual Knowledge Base Population - Cnam - Conservatoire national des arts et métiers
Communication Dans Un Congrès Année : 2024

Shadowfax: Harnessing Textual Knowledge Base Population

Résumé

Knowledge base population (KBP) from texts involves the extraction and organization of information from unstructured textual data to enhance or create a structured knowledge base. This process is crucial for various applications, such as natural language understanding, question-answering systems, and knowledge-driven decision-making. However the difficulty lies in the complexity of natural language, which is nuanced, ambiguous, and context-dependent. Extracting accurate and reliable information requires overcoming challenges such as entity disambiguation and relation extraction which are time-consuming tasks for users.Shadowfax is an interactive platform designed to support users by streamlining the process of knowledge base population (KPB) from text documents. Unlike other existing tools, it relies on a unified machine learning model to extract relevant information from unstructured text, enabling operational agents to gain a quick overview. The proposed system supports a variety of natural language processing (NLP) tasks using a single architecture, while presenting information in the most comprehensive way possible to the end user.
Fichier non déposé

Dates et versions

hal-04677434 , version 1 (26-08-2024)

Identifiants

Citer

Maxime Prieur, Cédric Du Mouza, Guillaume Gadek, Bruno Grilheres. Shadowfax: Harnessing Textual Knowledge Base Population. SIGIR 2024: The 47th International ACM SIGIR Conference on Research and Development in Information Retrieval, Jul 2024, Washington DC, United States. pp.2796-2800, ⟨10.1145/3626772.3657666⟩. ⟨hal-04677434⟩
20 Consultations
0 Téléchargements

Altmetric

Partager

More