Research
Academic Support Networks We investigate the thesis journey of PhD. students by extracting information from 27 thousand theses collected from pqdtopen.com (which is manually scraped from the website by a webscraping algorithm coded by me). The paper aims to inspect who is acknowledged, who is not, and how are they acknowledged using state of the art natural language processing tools that allow making large-scale research. We examine how the support providing entities are connected to each other by running network analysis and study gender based and disciplinary differences. We also show which factors influence productivity level of PhD students.
Covid-19 Vaccine Perception in Turkey In SICSS 2021, we started investigating the impressions of people tweeting in Turkish about Covid-19 vaccines. We study:
- The sentiment towards different vaccines and their nomenclatures.
- Media coverage of the media channels about vaccines.
- Bot detection.
- Usage of pronouns.
Two of our articles were published in Teyit.org (in Turkish) :
- How are Covid-19 vaccines perceived by society and the media?
- Vaccine disinformation in Turkey: Who is spreading false information, how is their network?
Covid-19 Perception & Dis/Misinformation Spread in Turkey In this paper, I present a model that takes into account both textual and non-textual characteristicsof social media users by utilizing graph convolutional neural networks along with a pre-trained Transformers model. I’ll use the suggested model to discover political traits of vulnerable groups on social media who are susceptible to misleading information. My goal is to assist policymakers in developing systematic techniques to reaching different stratifications of the general public.