Experiences
I am a first-year PhD candidate at Imperial college, under the supervision of Prof. Lucia Specia, and a member of the Natural Language Processing group. I am currently working on a project related to toxicity detection in online social networks, focussing on the importance of multimodality.
Qustodio is a SaaS Startup that provides software for parental control. I was part of the Product Team, responsible for providing analytics and insights regarding the product. More specifically:
- Develop and mature product success KPIs and SaaS metrics. Migrate all the analysis from python to ChartIO. Build and manage dashboards to drive awareness and understanding of key metrics.
- Partner closely with product teams to deliver analytics that inform product direction and strategy. Assist with analysis of A/B experiments designed by POs.
- Apply expertise in quantitative statistical analysis and data mining to create actionable insights.
I have participated to the European Research Marie Curie Programme ENCASE. During my secondment I worked on various research tasks, as listed below:
- Analysis of Hate speech in Social Media by Crowdsourcing and analysing a large-scale dataset of Twitter as well as creating an architecture for detection of hate speech using deep neural networks.
- Assisting the detection of controversy in online content, using traditional bayesian models.
- Quantitative and qualitative analysis on the dissemination of fake news in social media platforms.
As a member of the OSWINDS Research Group, I have participated in a number of research projects as listed below:
- Crowdsourcing opinions related to match fixing and creating a visualization tool to present the outcomes of the project, for the Erasmus+ european Research Programme, Fix the Fixing.
- Studying the impact of external resources on Twitter Information Diffusion process, in collaboration with Zayed University, Dubai.
- Outgoing fellowship with Telefonica, Spain, for the ENCASE European Research Programme.
Collection and clustering of geo-annotated data from FlickR, to discover and study underlying patterns.
Publications
- Founta, A. M., Chatzakou, D., Kourtellis, N., Blackburn, J., Vakali, A., & Leontiadis, I. "A unified deep learning architecture for abuse detection". In Proceedings of the 10th ACM Conference on Web Science (pp. 105-114). ACM. Boston (MA), USA.
- Founta, A. M., Djouvas, C., Chatzakou, D., Leontiadis, I., Blackburn, J., Stringhini, G., Vakali, A., Sirivianos, M., & Kourtellis, N. "Large Scale Crowdsourcing and Characterization of Twitter Abusive Behavior". Proceedings of the 12th International AAAI Conference on Web and Social Media (ICWSM '18), Stanford (CA), USA.
- Koursoumis, A., Gkatzou, E., Founta, A. M., Mavriki, V.I., Talvis, K., Mprilis, S., Aliwat, A.A. & Kermanidis, K. “Learning to Case-tag Modern Greek Text”. Proceedings of the 7th Hellenic Conference on Artificial Intelligence – SETN 2012. Hellenic Artificial Intelligence Society (EETN), Lamia, Greece.
Skills & Proficiency
Python
Machine Learning
Data Handling
Statistical Data Analysis
Front-end Web Development
AWARDS
I was awarded a scholarship for the presentation of a poster at the womENcourage 2017 scientific event. In order to receive the award, I had to carry out a few hours of volunteer work during the conference.
The scholarship was given to me as I ranked first between the candidates of the Postgraduate Studies Programme of the Informatics Department, AUTh. My responsibilities included being an assistant of the Operating Systems course as well as the Web Information Systems course of the Undergraduate Studies Programme of the Department. As an assistant I had to make some lectures, create and mark projects and create numerous self-assessments to help the students better understand each lecture.
The scholarship was given to me as I ranked first between the students of the Postgraduate Studies Programme of the Informatics Department, AUTh. My responsibilities included being an assistant of the Web Information Systems course as well as the Operating Systems course of the Undergraduate Studies Programme of the Department. As an assistant I had to make some lectures and arrange some guest lectures and create and mark some projects for both courses.
Projects
Here is a list of my projects and projects I have been a part of.
Crowd-sourcing in Social Media, including Twitter and Youtube, regarding match fixing and bet fixing as part of the ongoing Erasmus+ project FIX the FIXING: Proactive quelling of sports events manipulation (Collaborative Partnership). Tasks included:
- Perform social media mining regarding match fixing cases
- Enable automatic data visualization of the results
- Build a tool for sentiment analysis of short informal texts
- Develop a web platform for the presentation of the results