Panagiotis Mavridis

Postdoc at the Web Information Systems group of the Faculty of Engineering, Mathematics and Computer Science (EEMCS/EWI), Delft University of Technology.

My specialty is to create, research and deploy Web Data Driven Hybrid Human-Machine Systems. I am also a Visiting Researcher of the Vrij Universiteit of Amsterdam, where I join the User Centred Data Science (UCDS) group.

Current Research

Currently I am working on developing data-driven approaches with humans in the loop. For instance deploying hybrid human machine systems in order to characterise bias in news videos and articles shared in online news media. My research involves working with an interdisciplinary team of media scholars and political scientists where we focus on defining, capturing and quantifying bias from various online news big media data sources. To this end, we develop automatic and semi-automatic workflows that involve humans in the loop in order to assist news media scholars with capturing bias.

CapturingBias Project

This project aims at giving an automated solution involving humans in the loop to provide with useful analytics on identifying and analysing bias cues to help media scholars and then the audience to identify bias coming from different news sources. Please feel free to visit the project website for more details: <text> capturebias.eu </text>

Teaching

In addition to my research portfolio, I am currently actively helping with the teaching procedures in TU Delft such as: supervising final year master students, helping with the newly found Crowd Computing course of TU Delft (human computation tutorials) and help with coaching students in team and personal projects related to computer science and data science. <text> You can look at my recent crowdsourcing quickstart tutorial on Figure8 platform here.</text>

Education and Previous work

I graduated from the Electrical and Computer Engineering School of the National Technical University of Athens on 2013 (ECE National Technical University of Athens). After that, I also received a Research Master with double affiliation from IMT Atlantique (formerly known as Telecom Bretagne of Brest, France) and the University of Rennes 1 in Human Centered Computer Science. During my PhD i worked for DRUID team (IRISA institute) under the Supervision of <text> David Gross-Amblard </text> and <text>Zoltan Miklos</text> . I obtained my PhD in November 2017 focusing on quality improvement for task assignment methods in human computation. During my PhD, I developed and tested a user and task model based on a scalable (for current Crowdsourcing platforms such as Amazon Mechanical Turk and FigureEight) hierarchical skill model that gives a significant improvement in quality when used for task assignment. My research was mainly funded by a MESR funding of the French Ministry of Education and part of my research was also funded by the ANR HEADWORK project.

Awards

My commitment and dedication to research has been rewarded with a PhD grant from the French Ministry of Education and latter on I had the honour to receive the best student paper award for my paper that I presented in WWW2016 in Montreal, Canada.

More

For more information, please feel free to ask me for my latest CV and visit my LinkedIn profile.

Dr. P. Mavridis

Visiting Address
Building 28

Room: 4.E.220 East 4rd floor
Van Mourik Broekmanweg 6
2628 XE Delft
The Netherlands

Mailing Address
EEMCS, Web Information Systems
P.O. Box 5031, 2600 GA Delft
The Netherlands