Postdoc in Brazil combining NLP, complex networks and ML Full-time NEW

Predicting and analyzing success in science with complex networks and natural language processing

A postdoctoral position is open within the Núcleo Interinstitucional de Linguística Computacional (NILC), under the supervision of Prof. Dr. Diego Raphael Amancio, to investigate computational methods for forecasting the behavior of the scientific literature. The project combines machine learning, natural language processing (NLP) and complex networks to represent scientific knowledge as semantic networks and to predict emerging research topics, their potential impact, and the success of new combinations of concepts, a research line within the growing field of Science of Science.

The selected researcher will work on tasks such as link prediction in semantic networks, construction and analysis of co-occurrence and citation networks, vector representations of text (e.g. word2vec, BERT, RoBERTa, GPT-family models), structural and semantic node embeddings (e.g. node2vec, struct2vec), and impact metrics (citations, disruption index, structural centrality). Large-scale datasets such as OpenAlex, arXiv and the American Physical Society (APS) corpus will be used.

The position is hosted at the NILC, Institute of Mathematics and Computer Science (ICMC), University of São Paulo (USP), in São Carlos, SP, Brazil. The work also involves close collaboration with researchers at the Institute of Physics of São Carlos (IFSC-USP) and international collaboration. 

São Carlos, with around 220,000 inhabitants, is situated in the central region of the state of São Paulo, 240 km from the state capital and 850 meters above sea level. Due to its mild climate, the city is known as the “Brazilian Capital of the Climate”, and for its growing number of high-technology startups and companies it is also called the “Brazilian Capital of Technology”. São Carlos is the Latin American city with the highest ratio of PhDs per capita — roughly 1 per 100 inhabitants — and is the only Brazilian city of its size hosting two top universities, the University of São Paulo and the Federal University of São Carlos (UFSCar), as well as two Embrapa research centers.

Requirements

Besides a solid background and a recent PhD (obtained less than 7 years ago, per FAPESP rules) in computer science, applied mathematics, physics, electrical/computer engineering, data science or statistics, the candidate should have:

• Strong publication record in machine learning, NLP, complex networks and/or Science of Science, preferably in high-impact journals;
• Strong background knowledge in complex networks, graph neural networks, machine learning, NLP and statistics;
• Good knowledge of programming languages and tools, especially Python (and relevant ML / network science / GNN libraries);
• Very good knowledge of written and spoken English (Portuguese is not required);
• Strong teamwork capacity and willingness to engage in national and international collaborations.

Funding and conditions

The position is supported by a FAPESP Postdoctoral Fellowship of R$ 12,570.00 per month (tax-free), plus a yearly technical reserve (10% of the annual value of the fellowship) to be spent on research-related items, and installation support for candidates relocating to São Carlos. The fellowship lasts 24 months. As a reference, a good flat close to the University can be rented for around R$ 1,500.00 per month.

Starting date: as soon as possible, upon agreement.

How to apply

If you are interested and believe you qualify, please send a cover letter, a résumé with a list of publications, and the names, e-mail addresses and phone numbers of at least three references to Prof. Dr. Diego Raphael Amancio. Please mention ”Application to Postdoctoral position — Science of Science / NLP / Complex Networks” in the subject line of your email. 

Tagged as: Computer Science