Logik geht durch den Magen -- modeling interesting phenomena in (not only) food computing

Aus International Center for Computational Logic
Wechseln zu:Navigation, Suche

Logik geht durch den Magen -- modeling interesting phenomena in (not only) food computing

Vortrag von Weronika T. Adrian
Abstract: The emerging field of food computing tackles, among others, problems of knowledge acquisition, modeling, and processing within the culinary domain. Food-related knowledge is often implicit, contextual, or culture-dependent. Making at least part of it explicit, with some formalization, opens up possibilities to develop intelligent knowledge-based solutions to assist humans in preparing and optimizing food. An interesting and relevant problem is searching for substitutions in food recipes, a task motivated by different constraints and goals of a person, and leveraging the knowledge of dieticians and/or food technologists.

Addressing this challenge, the following questions emerge: What is relevant knowledge when it comes to food recommendations and AI-based decision support for ingredient substitution? Is there space for knowledge-based systems in the age of large language models? How to formally represent and reason about diets, allergies, and other restrictions? What food-related knowledge graphs and ontologies are there and are they useful for practical applications? In this seminar, I will share my experience from the TAISTI project (“Development of a Technology based on Artificial Intelligence for inferring SubsTitutable recipe Ingredients”, https://www.taisti.eu/), in particular ontology and knowledge graph engineering, and logical reasoning over integrated knowledge. Apart from practical challenges, I will also discuss interesting questions about the similarity of entities -- a problem that goes beyond the domain of food. I will present the entity set expansion problem and discuss recent proposals for formally describing the similarity of instances in knowledge bases and retrieving more entities ``of the same kind.

Speaker bio: Weronika T. Adrian, Ph.D. (wta@agh.edu.pl, http://wtadrian.eu, http://kraken.edu.pl) is an Assistant Professor and the Deputy Head of the Department of Applied Computer Science, Faculty of Electrical Engineering, Automatics, Computer Science and Biomedical Engineering, AGH University of Kraków, Poland. Her research interests cover the theory and practice of semantic technologies, knowledge graphs, and logic programming. She (co-)authored nearly 70 publications in the field of AI. Obtained her Ph.D. in Mathematics and Computer Science from the University of Calabria (Italy), followed by a 2-year post-doctoral contract in the group of Prof. Nicola Leone. Dr. Adrian has been involved in multiple national and international R&D projects and EU programs. She is a laureate of the “Top 500 Innovators” scholarship at Stanford University (USA). Founder of Creativity and Innovation Lab Foundation, a founder member of the Polish Artificial Intelligence Society, and a former member of IEEE.

The talk will take place in a hybrid fashion, physically in the APB room 3027, and online through the link:

https://bbb.tu-dresden.de/b/pio-zwt-smp-aus