KIT-IA (Knowledge-drIven Techniques for Intelligent Applications in heterogeneous contexts)

Introduction

Knowledge management plays a key role in Artificial Intelligence (AI). Since half a century ago, the use of knowledge in expert systems, semantic networks, or frames made it possible to solve complex real-world problems in many domains with the most popular examples being in medicine diagnostics. Since then, Knowledge Representation and Reasoning (KRR) is still one of the key subfields of AI, and numerous different strategies to manage knowledge have been proposed and successfully evaluated in practical applications. More modern approaches to represent knowledge include ontologies, Open Linked Data, or, more recently, knowledge graphs (KG). Those techniques were mainly developed by the Semantic Web (SW) community (hence they are usually called semantic technologies), but can also be applied to any non-Web-based application. Complex AI systems usually need to incorporate a KRR module in conjunction with other AI techniques such as Machine Learning (ML), Natural Language Processing (NLP), etc. to actually achieve their goals.

The main objective of this project is to propose novel knowledge-driven and multilingual-aware techniques improving the services offered by intelligent systems, regardless of the application domain. As examples of these techniques, we want to link the connected data to their actual meanings, being able to access them following the users intended meanings regardless their language, supporting flexible queries or queries expressed in natural language, extracting information in an intelligent way from unstructured sources, just to name few. Our techniques will address heterogeneous contexts and different knowledge domains, broadening thus the applicability of our proposals. To show the usefulness of our developments, we will ground them to some specific real-world scenarios as well.

In particular, we plan to address different research problems:

  1. Improvement of knowledge-based querying and information access. This includes extracting the actual structure of a KG, using graph embeddings to build ontology axioms, computing the semantic difference between ontologies, extending existing query answering systems to accept formal queries or natural language and to exploit KGs, developing novel methods for flexible query answering, etc.
  2. Building knowledge-driven Natural Language Processing (NLP). This includes combining semantic technologies and different language models (both non-contextualized, e.g., word2vec, and contextualized, e.g., transformers-based, ones), using KGs to model multilingual data and to improve translations or cross-lingual access to information, etc.
  3. Development of intelligent applications for mobile users. This includes adapting all the previous techniques to work on mobile devices, providing support for KGs on mobile devices, developing novel techniques for adaptive semantic reasoning on mobile devices, or applying the paradigm of Personal Knowledge Graphs to mobile computing.

Team members

Photo of the member Photo of the member Photo of the member Photo of the member Photo of the member Photo of the member Photo of the member
Eduardo Mena
(Main researcher 1) 		Fernando Bobillo
(Main researcher 2) 		Carlos Bobed Ignacio Huitzil Jorge Bernad Jorge Gracia Lacramioara Dranca

Collaborators

Related publications: journals

Related publications: conferences

Last update: 12/11/2024

Back to previous page

Última actualización: 12/11/2024

Regreso a la página anterior