Unit:
Κατεύθυνση Μεγάλα Δεδομένα και Τεχνητή ΝοημοσύνηΠληροφορική
Supervisors info:
Μανόλης Κουμπαράκης, Καθηγητής, Τμήμα Πληροφορικής και Τηλεπικοινωνιών, ΕΚΠΑ
Μέμα Ρουσσοπούλου, Καθηγήτρια, Τμήμα Πληροφορικής και Τηλεπικοινωνιών, ΕΚΠΑ
Αλέξανδρος Ντούλας, Επίκουρος Καθηγητής, Τμήμα Πληροφορικής και Τηλεπικοινωνιών, ΕΚΠΑ
Original Title:
A Greek Question Answering System over Knowledge Graphs
Translated title:
A Greek Question Answering System over Knowledge Graphs
Summary:
In the rapidly evolving fields of information retrieval and artificial intelligence, question answering systems have become increasingly popular and have seen significant advancements over the years. These systems provide an intuitive method for querying structured
data sources allowing users to ask questions in natural language and receive accurate
and relevant answers. Utilizing the power of Knowledge Graphs users can get the direct
answer rather than a list of potential results like popular search engines used to do.
While Knowledge Graph Question Answering systems have seen an increase in English
and other major languages, there is a lack of support for the Greek language. Through this
thesis, we aim to develop a KGQA system specifically designed for the Greek language.
Our focus for our pipeline is able to answer generic but simple questions over Wikidata
Knowledge Graph. Our proposed method is designed to be modular in order to support
additional open-source Knowledge Graphs.
Main subject category:
Technology - Computer science
Keywords:
Question Answering, Knowledge Graphs, Wikidata
