Machine Learning Applications in Science-of-Science Research
| Témavezető: | Béres Ferenc |
| SZTAKI, Informatikai Kutatólaboratórium | |
| email: | beres@sztaki.hu |
Témavezetők
- Benczúr András (SZTAKI)
Projekt leírás
The rapidly growing volume of scholarly publications provides new opportunities to study how scientific ideas emerge, spread, and achieve long-term impact. Science-of-science research applies computational, statistical, and network-based methods to answer questions concerning publications, citations, researchers, institutions, collaborations, and research topics.
The objective of this project is to develop and evaluate machine-learning methods for a selected science-of-science problem. Possible directions include predicting the long-term impact of publications, identifying influential or award-winning papers, analyzing citation trajectories, measuring scientific novelty, detecting emerging research areas, or studying patterns of collaboration and institutional influence.
The student will construct a reproducible data-analysis pipeline using an open scholarly database such as OpenAlex or Semantic Scholar. Depending on the selected research question, the project may combine bibliometric indicators, citation networks, author and institutional features, and text or document embeddings. Suitable models range from interpretable statistical baselines to tree-based methods, neural networks, graph algorithms, or pretrained models for tabular and scientific-text data.
Előfeltételek
- Working knowledge of Python and common data-analysis libraries such as NumPy, pandas, Matplotlib, and scikit-learn.
- Ability to work independently with technical documentation and scientific literature.
- Experience working with complex datasets or agentic coding frameworks is a plus
Hivatkozások
Y. Dong, R. A. Johnson, and N. V. Chawla, “Can scientific impact be predicted?” IEEE Transactions on Big Data, 2016.
J. Priem, H. Piwowar, and R. Orr, “OpenAlex: A fully-open index of scholarly works, authors, venues, institutions, and concepts,” arXiv:2205.01833, 2022.
R. M. Kinney et al., “The Semantic Scholar Open Data Platform,” arXiv:2301.10140, 2023.
A. Cohan, S. Feldman, I. Beltagy, D. Downey, and D. S. Weld, “SPECTER: Document-level representation learning using citation-informed transformers,” arXiv:2004.07180, 2020.
A. Singh, M. D'Arcy, A. Cohan, D. Downey, and S. Feldman, “SciRepEval: A multi-format benchmark for scientific document representations,” in Proceedings of EMNLP, 2022.
R. Vella, A. Vitaletti, and F. Silvestri, “Predicting award-winning research papers at publication time,” arXiv:2406.12535, 2024.