Research Group
Ivan Srba
Research areas: artificial intelligence, machine learning, data analysis, disinformation and misinformation, user behavioral data, social computing, collective intelligence
Position: Researcher
Ivan has expertise in artificial intelligence, machine learning and data analysis, with a focus on data emerging from user behavior on social computing and collective intelligence systems, such as social networks, knowledge sharing systems and e-learning. His current primary research interest is in countering online disinformation and misinformation.
He has co-authored more than 30 publications accepted to various international fora, including: 6 current-content high-impact journal manuscripts (one of which has received more than a hundred citations so far), 6 full conference papers (accepted to premiere conferences, such as RecSys) and 3 book chapters. These publications have received more than 640 citations in total, resulting in h-index 12 (according to Google Scholar, as of 2023/02). He has been involved in 12 research projects. His international research record includes projects with the Harvard University (USA), The University of Sheffield (UK), The Centre for Research & Technology, Hellas – CERTH (Greece), or the University of Lugano (Switzerland). He is currently a member of CEDMO project executive board and a task leader at HE RIA vera.ai project. He is a member of ACM educational and scientific computing society, Slovak.AI and a former member of the Datalys@PeWe research group. He regularly serves as a reviewer for international conferences and journals published by ACM, IEEE, Springer and Elsevier. So far, he has supervised 23 bachelor and 19 master successfully-defended theses.
He has participated as a data scientist and project lead in many research and industry projects. He was responsible for development of the first university-wide educational CQA system named Askalot, which has been deployed at three universities in three European countries and at MOOC system edX and which was used by 8000 students from 147 countries around the world. His other projects include a recommender system for a large internet shop (ZlavaDna.sk), a smart parking system (zaparkuj.to) and a multibank application developed in co-operation with a renowned European IT company (Unicorn) and a Slovak commercial bank (VUB).
News
- The research team, which I supervised with a big honor, has won the first place for detection of persuasion techniques in 6 out of 9 languages within the SemEval Task 3 Subtask 3 data challenge competition (our team name was KInIT). For the remaining languages, we placed in 2nd, 3rd and 4th places, and thus becoming univocally the most successful team! The task was addressed as a complex multilabel classification with 23 classes and with 3 surprise languages (not seen in any training data).
Selected achievements
- Best Paper award at the RecSys 2021 conference (the awarded full paper)
- Member of the excellence team PeWe of Slovak University of Technology in Bratislava, lead prof. Mária Bieliková
2020, Slovak University of Technology - Award for Supervision of the Best Master Thesis
Nov 2019, Tatra banka Foundation - Slovak Student Person of the Academic Year 2015/16 in Category Informatics, Mathematics and Physics
Dec 2016, Junior Chamber International – Slovakia - Rector’s Award for Excellent Study Results during Doctoral Study
Sep 2016, Slovak University of Technology - Rector’s Award for the Best Student of Doctoral Study for the Academic Years 2012/13, 2013/2014, 2014/2015
Slovak University of Technology - Best Master Thesis Dean’s Award
Nov 2012, Faculty of Informatics and Information Technologies, Slovak University of Technology - Rector’s Award for the Best Student of Master Study for the Academic Years 2010/11, 2011/12
Slovak University of Technology
Selected activities
- Invited talk on “Algorithmic auditing of political biases in recommender systems” at the online e-Symposium on Computational Politics 2023, Part of the AI Mellontology Symposium series (March 1, 2023)
- Participation at the panel discussion on Technology and AI at the CEDMO International Conference #CIC2022 (September 22-23, 2022, Prague)
- Invited presentation at Hackathon on Code against Hate, 2020
- Invited presentation at RuleML webinar, 2020
- Attendance at Dagstuhl seminar “Cybersafety Threats – from Deception to Aggression”, 2019
- Attendance at Heidelberg Laureate Forum, 2017
Professional Service
As a reviewer for journals:
- ACM Transactions on Information Systems (ACM)
- ACM Transactions on Computer-Human Interaction (ACM)
- ACM Transactions on Management Information Systems (ACM)
- Computer Science Review (Elsevier)
- Technology, Knowledge and Learning (Springer)
- Data Mining and Knowledge Discovery (Springer)
- IEEE Transactions on Learning Technologies (IEEE)
- Computers & Education (Elsevier)
- a program committee member for conferences
As a program committee member for international conferences:
- Conference on User Modeling, Adaptation and Personalization (UMAP)
- International Joint Conference on Artificial Intelligence (IJCAI)
- Conference of the European Chapter of the ACL (EACL)
As a member of professional societies:
Selected Projects
Misinformation Detection in Healthcare Domain
APVV SK-IL-RD-18-0004. 2018-2020, Partner: Bar Ilan University (prof. Sarit Kraus), Bielikova, M. – principal investigator
Innovative Methods of Education and Infrastructure for Informatics and Software Engineering
SCOPES program, Swiss National Science Foundation. 2015-2018, Bielikova, M. – principal investigator for FIIT STU
Selected Publications
An Audit of Misinformation Filter Bubbles on YouTube: Bubble Bursting and Recent Behavior Changes
Tomlein, M., Pecher, B., Simko, J., Srba, I., Moro, R., Stefancova, E., Kompan, M., Hrckova, A., Podrouzek, J., Bielikova, M. The negative effects of misinformation filter bubbles in adaptive systems…
A Comprehensive Survey and Classification of Approaches for Community Question Answering
Srba, I., Bielikova, M. Abstract: Community question-answering (CQA) systems, such as Yahoo! Answers or Stack Overflow, belong to a prominent group of successful and popular Web 2.0 applications, which are…
Selected Student Supervising
PhD
- Branislav Pecher – Addressing small labeled data in machine learning training. Started 2021. Co-supervised with Maria Bielikova.
Master
- Matej Čief – Use of Semi-supervised Learning for Antisocial Behavior Detection. Defended 2021
- Peter Mačinec – Fake News Detection Using Semi-supervised Learning. Defended 2021
- Veronika Žatková – Health related fake news detection using data analysis and machine learning. Defended 2020
- Štefancová Elena – Recommendation Taking the Time Aspects of Users and Items into Account. Defended 2019
- Mačina Jakub – Recommendation of New Questions in Online Student Communities. Defended 2017
- Grznár Marek – Adaptive Collaboration Support in Community Question Answering. Defended 2015
Bachelor
- Mačinec Peter – Fake news detection using machine learning. Defended 2019
- Babinec Peter – Automatic Tag Recommendation in CQA Systems. Defended 2017
- Huňa Adrián – User Reputation in Community Question Answering. Defended 2015