Hello, I'm Giannis, a passionate Data Scientist with a keen interest in weather data analysis. My expertise lies in utilizing a combination of rule-based and machine learning methods to craft cutting-edge post-processing weather models. I love delving into weather data, uncovering hidden insights, and contributing to the progress of meteorology.
In July 2023, I completed my studies at the Department of Informatics and Telecommunications of the National and Kapodistrian University of Athens. My specializations included Data Management and Knowledge, as well as Software. You can find my published thesis, "Machine Learning Snowfall Retrieval Algorithms for Satellite Precipitation Estimates," here.
In 2019, I graduated from High School, where my academic achievements included qualifying for the National Mathematical Olympiad, organized by the Hellenic Mathematical Society. Additionally, I actively participated in the Panhellenic Competition in Informatics.
Starting from November 2022, I've been working as a Data Scientist at WeatherXM, a community powered web3 weather station network. My role centers on analyzing weather data, evaluating and enhancing forecasts, implementing fault data detection mechanisms, and creating impactful weather data visualizations.
I am currently contributing as an independent consultant for the EuroHyPerCon project, which focuses on the connectivity requirements and network design for High-Performance Computing (HPC) systems in Europe.
I am currently contributing as an independent contractor for the development of a Climate Risk Assessment Platform.
I have previously developed a weather website and volunteered as a Teaching Assistant for the "Introduction to Programming" and "Data Structures" courses at the Department of Informatics and Telecommunications of NKUA, supporting students in their learning process. In addition, I have volunteered as a translator for the CERN Indico project and meteonetwork.it, contributing to the dissemination of valuable scientific content.
In December 2023 I was the Lead Instructor for the "Data Science and Machine Learning for Weather and Climate" workshop at Google DevFest Athens 2023.
I. Dravilas, S. Dafis, G. Kyros, K. Lagouvardos, and M. Koubarakis, 'Towards a Machine Learning Snowfall Retrieval Algorithm for GPM-IMERG', Environmental Sciences Proceedings, vol. 26, no. 1, 2023.
Also presented at the 16th International Conference on Meteorology, Climatology, and Atmospheric Physics on September 2023.
S. Keppas, H. Balis, I. Dravilas, and J. Pagonis, 'Designing a Global Weather Station Network', EGU General Assembly 2024, Vienna, Austria, 14-19 Apr 2024, EGU24-9241.
Will be presented at EGU General Assembly 2024.
During my thesis, I explored the application of advanced Machine Learning techniques, with a particular focus on Deep Learning algorithms, to estimate the phase of satellite precipitation estimates. Both numerical and in-situ observational weather data were also utilized in this study.
Over the past few years, I've accomplished a wide array of computer science projects, ranging from personal endeavors to university assignments.
Development of a high-resolution, post-processing, rule-based snow forecasting model for Greece.
Sentiment Classification of users' text reviews using Deep Learning (Feed-forward, Recurrent neural networks and pre-trained BERT model) and development of a Question Answering engine.
Development of a waterspout forecasting model for the Mediterranean Sea, using SWI Index and numerical weather data.
Classification of short music clips into different music genres using Feedforward and Convolutional Neural Networks.
Weather data visualisations in python3, using data in GRIB, GRIB2, netCDF formats and Matplotlib, Cartopy libraries.
Facial recognition using Yale B database face images, employing the Eigenfaces technique encompassing PCA and a nearest neighbour classifier.
Classification of images containing handwritten digits using SVM.
Sentiment Analysis of Covid-19 tweets fetched from Twitter, using various Data Mining techniques and Machine Learning algorithms.
Data Analysis of New York City taxi trips and creation of a Random Forest Regressor for predicting the duration of taxi trips.
Implementation of polygonization algorithms for point sets in a plane and area optimization using operational research concepts in C++ with CGAL library.
Development of software and protocols in C/C++ for Linux, using processes, threads, pipes, signals and the internet protocol.
AI-based solution for the contraint satisfaction Radio Link Frequency Assignment Problem.
Development of parallel solutions for Monte Carlo simulations using Pthreads, OpenMP and MPI.