Nilay Tüfek Özkaya

Date/Time: 12.02.2025, 12:00 

Location: D2.2.094 

Abstract 

This presentation will explore the integration of natural language-based technical documents, ontologies, knowledge graphs (KGs), large language models (LLMs), and natural language processing (NLP). Specifically, it will cover the following key aspects:

(1) Semantic Validation: How domain-specific technical documents can be analyzed to extract relevant information and transform it into compliance rules for validating semantic artefacts.

(2) Knowledge Retrieval: Using (competency) questions to query knowledge graphs and retrieve structured insights efficiently.

(3) Model Extension: How knowledge graphs can be leveraged with LLM-driven natural language queries (NLQ) to extend information models dynamically.

(4) RAG Solutions on KGs: Developing a Retrieval-Augmented Generation (RAG) approach to enable large-scale knowledge graph integration with LLMs.

The proposed framework aims to enhance knowledge extraction, validation, and augmentation in the OPC UA standard domain, contributing to more efficient and intelligent information management.

Bio 

Nilay Tüfek Özkaya graduated with a double major in Electronics Engineering and Computer Engineering from Istanbul Technical University (ITU) in 2013. She completed her master's degree at ITU and spent time at the University of Manchester as a visiting AI researcher. With over 12 years of experience at Siemens as an AI Researcher, she is currently pursuing her PhD at TU Wien. Her research focuses on artificial intelligence, natural language processing (NLP), large language models (LLMs), and knowledge graphs.