Academic Work

Theses, research projects and policy writing

Completed academic outputs, each labelled by type, authorship and status. None of these items is a peer-reviewed publication.

Theses

Individually authored degree theses. Full structured summaries are on the Research page.

2024

Master's Thesis

Investigating the Multifaceted Factors Influencing Information and Communications Technology (ICT) Growth

MSc in International Economy and Business · Corvinus University of Budapest

Authorship: Sole author

A cross-country panel study that treats ICT growth, measured by the share of the population using the internet, as an outcome of social, economic and political conditions. Using country and year fixed-effects models on World Bank and V-Dem data, it finds that GDP per capita and FDI inflows are positively associated with internet use, that the association with educational inequality is positive in the global model but varies by region, and that the negative association with political competition is sensitive to model specification.

Methods and topics
  • Country and year fixed-effects panel models
  • Clustered robust standard errors
  • Alternative and dynamic specifications
  • Regional and period-specific analysis
Key findings
  • GDP per capita and FDI net inflows were positively associated with internet usage.
  • Educational inequality had a positive coefficient globally, with signs that varied by region.
  • The negative association with political competition was not robust to dynamic and period-specific models.
Keywords
  • ICT
  • digital inclusion
  • educational inequality
  • political competition
  • FDI
  • economic development
  • internet penetration
Status
  • Completed
  • Associational design — not a causal analysis

2019

Bachelor's Thesis

Opinion Mining Using Naive Bayes Algorithm on Online Accommodation Customer Reviews

Bachelor of Computer Science · Universitas Gadjah Mada

Authorship: Sole author

An opinion-mining system that classifies English-language Airbnb reviews as positive, negative or neutral. TF–IDF weighting and a Naive Bayes classifier were implemented from scratch in Python and evaluated with 5-fold cross-validation. Keeping negation words such as “not” and “no” during stop-word removal raised average accuracy from 69% to 72% and F-measure from 70% to 73%.

Methods and topics
  • Text preprocessing
  • TF–IDF weighting (implemented from scratch)
  • Naive Bayes classification (implemented from scratch)
  • 5-fold cross-validation
Key findings
  • Average accuracy 72%, precision 75%, recall 72% and F-measure 73% with modified stop-word handling.
  • Neutral reviews were the hardest class to identify.
Keywords
  • opinion mining
  • Naive Bayes
  • TF–IDF
  • text classification
  • customer reviews
Status
  • Completed

Research Projects

Research carried out as part of taught courses. These are not peer-reviewed publications.

2023

Course Research Project

An Empirical Analysis of Factors Impacting on Political Stability

Corvinus University of Budapest · Course: Data Analysis for World Economics

Authorship: Co-authored group project

A co-authored course project using country-level panel data on 202 countries to study how economic, educational, political and institutional factors relate to political stability. Fixed-effects models with population weights and robust standard errors found positive associations with GDP per capita, domestic autonomy and educational equality, and negative associations with suffrage and coup attempts, with differences across regions.

Methods and topics
  • Country-level panel data
  • Fixed-effects regression with population weights
  • Robust standard errors
  • Model comparison, regional analysis and robustness checks
Keywords
  • political stability
  • panel data
  • fixed effects
  • institutions
  • V-Dem
Status
  • Course project
  • Co-authored
  • Not peer reviewed

Policy and Research Communication

Policy writing and research communication produced for coursework.

2023

Course Policy BriefResearch Poster

Impact of ChatGPT in Education: How Can We Balance the Benefits and Risks of ChatGPT?

Corvinus University of Budapest · Course: Research Design and Methodology I

Authorship: Co-authored course project

A co-authored policy brief assessing the educational benefits and risks of ChatGPT. The group combined a literature review, comparative policy analysis, case studies and a questionnaire with 61 respondents, and developed policy recommendations. An accompanying research poster was co-presented as the final course deliverable.

Methods and topics
  • Literature review
  • Comparative policy analysis
  • Case studies
  • Questionnaire (61 respondents)
  • Policy recommendations
Status
  • Course project
  • Co-authored
  • Not peer reviewed
  • Poster presented within the course, not at an external conference

Working papers and publications

Future working papers, preprints and publications will be listed here.