Name: Alexis Filippakopoulos
Type: User
Company: Athens University of Economics and Business
Bio: Undergrad @ Informatics Department, Athens University of Economics and Business.
Location: Athens, Greece
Alexis Filippakopoulos's Projects
Implementing a Personalized Federated Learning and Split Learning hybrid framework 'SplitGP' as part of my bachelor thesis.
Using CNN keras sequential model to classify dog and cat images.
A game created in C++ named Spacewars which is an arcade survival game with spaceships.
EasyWash is a user interface for washing machines designed to ameliorate the user experience of visually impaired people. The interface leverages techniques of computer vision, speech recognition and natural language processing to deliver a user-friendly experience for the specified target group.
Model developed using Python's keras that classifies if a person is wearing a mask or not.
The backend code for a fully distributed messaging app in Java. The system implements the Publish - Subscribe architecture, where the are 3 brokers (servers) where clients are connected to and can talk to each other, upload stories, make friends and many more. The system is also fault tolerant.
The frontend code for a fully distributed messaging app made with android studio. The system uses the Publish - Subscrib architecture, where the are 3 brokers (servers) where clients are connected to and can talk to each other, upload stories, make friends and many more. The system is also fault tolerant
Detecting the colors inside an image using Python (pandas, cv2).
Implementing Naive Bayes and Logistic Regression algorithms from scratch in order to perform sentiment analysis on the IMDB dataset.
Partly replicating a paper investigating the 1978 Argentinian World Cup, taking place under the argentinian jounta and the change in repression events as the tournament was closing in.
Working on an artificial tinder messages dataset to extract frequent message sequences and visualize the conversation endings (last 6 messages sequence).
A simple multithreaded motion detection alarm system using python's cv2 library.
Partly replicating two studies carried out on social capital. The studies appeared in the journal Nature.
PyTorch implementation of SimCLR: A Simple Framework for Contrastive Learning of Visual Representations by T. Chen et al.
Predicting the disruptiveness of a paper as proposed by the referenced study.
Multithreaded code in C to reserve seats in a theater.
Classifying weather images into categories: Cloudy, Foggy, Rainy, Shine, Sunrise using Python.