- This event has passed.
SWC Research @Semantics 2022
September 13, 2022 - September 15, 2022
Applying Semantic AI platforms for Augmented Intelligence
Presenter: Robert David, Patrice Neff
Wednesday, September 14, 2022 – 10:30 to 12:00
In recent years, augmented intelligence systems have advanced to provide humans with insights for knowledge discovery and decision making. Using a cooperative approach, these systems use powerful artificial intelligence (AI) to augment human cognitive activities. To do so, the AI capabilities have to complement human actions and therefore be explainable and controllable, while still providing implicit insights based on statistical models. Semantic Artificial Intelligence (Semantic AI), as the combination of machine learning and knowledge models, provides such a methodology.
Poster: Proposal for PORQUE, a Polylingual Hybrid Question Answering System
Presenter: Anna Breit
Wednesday, September 14, 2022 – 17:45 to 18:30
Semi-automated Generation of Multilingual Domain-specific Taxonomies
Presenter: Martin Kaltenböck, Ilan Kernermann
Thursday, September 15, 2022 – 10:15 to 11:15
The use of and need for domain-specific taxonomies are rapidly augmenting, as part of the growing interest in multilingual knowledge management. However, creating a fine taxonomy for any domain or industry use case can require many months of intensive expert work. To face this challenge, we converge human created and curated data with automated processes and a Knowledge Graph tool. Namely, we apply cross-lingual lexicographic content from KD-Lexicala with Semantic Web Company’s PoolParty Semantic Suite to generate multilingual domain-specific taxonomies.
Learning ontology classes from text by clustering lexical substitutes derived from language models
Presenter: Anna Breit
Thursday, September 15, 2022 – 15:00 to 16:00
Many tools for knowledge management and the Semantic Web presuppose the existence of an arrangement of instances into classes, i. e. an ontology. Creating such an ontology, however, is a labor-intensive task. In this paper, we present an unsupervised method to learn an ontology from text. We rely on pre-trained language models to generate lexical substitutes of given entities and then use matrix factorization to induce new classes and their entities.