Mizeli Chaido (Phd Candidate)

Thesis title: “Extracting Knowledge Graphs from text through Natural Language Processing using Machine Learning”

Supervisor: Goulianas Konstantinos

Advisory Committee Members:

Konstantinos Diamantaras, Professor, Dept. of Information and Electronic Engineering, IHU
Bratsas Charalampos, Assistant Professor, Dept. of Information and Electronic Engineering, IHU

Abstract:

Publishing is a common form of communication and transaction on the Internet, with natural language serves as the main method.For instance, a vast quantity of digital data regarding individuals, companies, and even government units are generated every day in the field of digital public administration. This data is significant in terms of its semantic and conceptual value. In Natural Language Processing (NLP), Knowledge Graphs (KGs) facilitate the inference of new data and the integration of understanding into systems. Machine Learning (ML) has the potential to improve the performance and result accuracy.There are open research issues in the scientific field of ML related to NLP-based KG construction. The aim of the research is to introduce ML techniques for the extraction of KGs from text at the different NLP stages, to collect the data to produce specific knowledge on a range of topics. Developing methodologies to support the Greek language is another objective.The combined methodology of the above-mentioned scientific domains adds value to the proposed study.