Word Sense Disambiguation, Target Sense Verification, and Entity Linking
This line of work looks into challenges related to recognizing words in a text as specific entities, understanding their types and linking them to a knowledge graph.
ptlm_wsid (this link opens in a new window) by semantic-web-company (this link opens in a new window)
Usage of pre-trained language model to induce and disambiguate entity senses
by Artem Revenko, June, 2021
Project Demo: Legal Knowledge Graph for Multilingual Compliance Services
by Anna Breit, June, 2021
by Anna Breit, June, 2022
How to correctly link entities from text to your Knowledge Graph using Word Sense Induction and Disambiguation.
A quicker and more robust Word Sense Induction and Disambiguation approach for Entity Linking.
Enabling reliable disambiguation without the need to induce at all, even with incomplete sense inventories.
Introduction to Target Sense Verification disambiguation task formulation and WiC-TSV dataset.
Testing Domain-Adaptation Capabilities of Disambiguation Models on WiC-TSV-de, a German dataset for Target Sense Verification.
How Target Sense Verification improves domain-specific and enterprise disambiguation settings.
The main objective of Lynx is to create an ecosystem of smart cloud services to better manage compliance, based on a legal knowledge graph (LKG) which integrates and links heterogeneous compliance data sources including legislation, case law, standards and other private contracts.
Visit Project Website
The FFG-funded OBARIS project aims to advance the status quo in the area of auditable semantic artificial intelligence systems (SAISs), by investigating both conceptual aspects of these systems as well as a technology stack that facilitates transposing these system types into concrete settings.
Visit Project Website