George Katsogiannis

Athens, Greece

I am a PhD Student at Université Grenoble Alpes (UGA) in co-supervision by UGA and the Athena Research Center (ARC), and a research associate at ARC, under the supervision of Dr. Georgia Koutrika and Prof. Sihem Amer-Yahia. My work focuses on using Deep Learning and Natural Language Processing techniques to make data open and accessible to users without technical knowledge and is part of the INODE and FAIR Core 4 EOSC projects, funded by the EU. During the past few years, my work has resulted in 5 tutorials presented at major conferences [EDBT’21, SIGMOD’21, TWC’22, WSDM’23, VLDB’23], one query explanation system [eQsplain] that has been integrated into the INODE project, and a survey studying the fast-paced research area of deep learning text-to-SQL [VLDBJ’23].

My MSc degree was in Data Science and Information Technologies, with a specialization in Artificial Intelligence, at the University of Athens. My MSc thesis is titled “Data Democratisation with Deep Learning: Structured Query Translation from and to Natural Language”. Before that, I was studying towards my BSc degree in Computer Science at the University of Athens, where I completed my BSc thesis, titled “Translating Natural Language to SQL using Deep Learning”.

news

Nov 20, 2023	ScienceBenchmark has been accepted at VLDB’24! This is a joint work with ZHAW on Text-to-SQL benchmarking and data augmentation for complex scientific DBs.
May 25, 2023	Our tutorial (with M. Xydas and G. Koutrika) titled “Natural Language Interfaces for Databases with Deep Learning” has been accepted at VLDB’23!
Apr 11, 2023	I successfully defended his MSc thesis on “SQL Translation from and to Natural Language”!
Mar 1, 2023	Our vision paper (with C. Tsapelas and G. Koutrika) on NL Data Interfaces was adccepted at the BigVis Workshop, held in conjunction with EDBT/ICDT 2023.
Nov 7, 2022	Our tutorial, titled “Data Democratisation with Deep Learning: The Anatomy of a Natural Language Data Interface”, has been accepted at WSDM’23!

selected publications

VLDBJ
A survey on deep learning approaches for text-to-SQL

George Katsogiannis-Meimarakis, and Georgia Koutrika

The VLDB Journal, 2023

Abs Bib HTML

To bridge the gap between users and data, numerous text-to-SQL systems have been developed that allow users to pose natural language questions over relational databases. Recently, novel text-to-SQL systems are adopting deep learning methods with very promising results. At the same time, several challenges remain open making this area an active and flourishing field of research and development. To make real progress in building text-to-SQL systems, we need to de-mystify what has been done, understand how and when each approach can be used, and, finally, identify the research challenges ahead of us. The purpose of this survey is to present a detailed taxonomy of neural text-to-SQL systems that will enable a deeper study of all the parts of such a system. This taxonomy will allow us to make a better comparison between different approaches, as well as highlight specific challenges in each step of the process, thus enabling researchers to better strategise their quest towards the “holy grail” of database accessibility.
@article{katsogiannis2023survey, author = {Katsogiannis-Meimarakis, George and Koutrika, Georgia}, title = {A survey on deep learning approaches for text-to-SQL}, journal = {The VLDB Journal}, year = {2023}, issn = {0949-877X}, doi = {10.1007/s00778-022-00776-8}, url = {https://doi.org/10.1007/s00778-022-00776-8}, }
SIGMOD
A Deep Dive into Deep Learning Approaches for Text-to-SQL Systems

George Katsogiannis-Meimarakis, and Georgia Koutrika

In Proceedings of the 2021 International Conference on Management of Data, 2021

Abs Bib HTML

Data is a prevalent part of every business and scientific domain,but its explosive volume and increasing complexity make data querying challenging even for experts. For this reason, numerous text-to-SQL systems have been developed that enable querying relational databases using natural language. The recent advances on deep neural networks along with the creation of two large datasets specifically made for training text-to-SQL systems, have paved the path for a novel and very promising research area. The purpose of this tutorial is a deep dive into this area, covering state-of-the-art techniques for natural language representation in neural networks,benchmarks that sparked research and competition, recent text-to-SQL systems using deep learning techniques, as well as open problems and research opportunities.
@inproceedings{10.1145/3448016.3457543, author = {Katsogiannis-Meimarakis, George and Koutrika, Georgia}, title = {A Deep Dive into Deep Learning Approaches for Text-to-SQL Systems}, year = {2021}, isbn = {9781450383431}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3448016.3457543}, doi = {10.1145/3448016.3457543}, booktitle = {Proceedings of the 2021 International Conference on Management of Data}, pages = {2846–2851}, numpages = {6}, keywords = {deep learning, text-to-SQL}, location = {Virtual Event, China}, series = {SIGMOD/PODS '21}, }