Prof. dr. Pinar Yolum Birbil

Minnaertgebouw
Leuvenlaan 4
Kamer 3.07
3584 CE Utrecht

Prof. dr. Pinar Yolum Birbil

Hoogleraar
Responsible AI
030 253 4094
p.yolum@uu.nl

Current Projects:

 

Projects Completed at Bogazici University:

The project developed methods to detect and prevent privacy violations in online social networks. An ontology for online social networks has been developed to accommodate semantic reasoning. The developed techniques have been implemented in an open-source software and have been demonstrated in AAMAS and ECAI.

  • TUBITAK Project, Researcher. CaReRa: Content-Based Case Retrieval in Radiological Databases. 2011–2014. Completed.

The project developed a semantic search engine to search for medical cases in radiological databases. With consultation with radiologists, an ontology of liver for radiology (ONLIRA) has been designed and developed. The ontology has been used to annotate radiology images.

  • TUBITAK Project, Researcher. MANGO: MultiAgent Environment for Global Optimization. 2008–2011. Completed.

The project designed the principles for solving global optimization problems with software agents. A fully distributed environment has been developed that allowed independent developers to design agents to be plugged into the system. The project demonstrated that with various cooperation strategies multiagent systems can solve global optimization problems successfully.

 

The project developed various techniques to represent services semantically and developed algorithms to enable service evolution and composition.

  • TUBITAK Project, Researcher. Sign-Language Tutor. 2007–2009. Completed.


The broad aim of this project was to design a software to teach sign-language based on capturing users’ demonstration of signs and giving feedback based on video captures. As part of this project, we built a multiagent system that could enable signs to be shared over multiple terminals and enable group decision making on the correctness of signs.

 

Afgesloten projecten
Project
Using Multi-Agent Systems for Simulating Online Social Network Influence on Urban Planning 01-11-2021 tot 30-06-2022
Algemene projectbeschrijving

The widespread use of Twitter, Facebook and other social media platforms has increasingly influenced planning practices. Social media have been widely employed by governments as a “smart” tool for citizen engagement and participatory governance as well as for harnessing collective actions. However, there is a lack of study on what kinds of key factors influence the interactions and actions of various actors in social networking sites and how online social networks affect planning outcomes. This project aims to develop a multi-agent model to simulate online social network influence on participatory planning in the context of sustainable urban transition.This transdisciplinary study will bridge the knowledge gap between urban planning and computer science regarding online social network study. First, this study will develop a new methodology by using data collected from online social networks of selected case studies to adjust the parameters of  the multi-agent model to reflect the understanding and sentiment of online participants. The multi-agent model developed here will be implemented in a simulation environment, which will also serve as a platform for future studies. Second, it will investigate the complex role of social networking sites in contestation, empowerment and inclusion in urban sustainability transition. It will explore what extent social network sites increase the power of citizens, enhance the effectiveness of citizen participation, and facilitate social inclusion. Answering these questions will bring us one step closer to our goal of establishing social sustainability.

Rol
Uitvoerder
Financiering
1e geldstroom Seed money of “Urban Sustainability Transformations” Research Hub “Transforming Cities”
Projectleden