About this Project
BiblioAssistant is an automated system designed for keeping track of new scientific literature and summarizing research articles in the fields of hydrology, climate, and meteorology. The selection is specifically tailored to the research interests of the Hydrology and Climate Change research group at the Ebro Observatory.
The project is open source and the code is publicly available on GitHub.
How it Works
The system operates through a specialized pipeline:
- Ingestion: It monitors scientific literature using the OpenAlex API. We are deeply grateful to OpenAlex for providing free, open access to their massive database of scholarly works.
- Discovery & Learning: The system proactively searches for new papers based on keywords, citations of existing research, and specific journals. Notably, BiblioAssistant self-learns: if a journal or an author appears in at least three relevant papers, the system automatically adds them to its permanent watchlist.
- Relevance Filtering: A Large Language Model (LLM) analyzes titles and abstracts to identify papers relevant to the specific research interests of the Hydrology and Climate Change team at the Ebro Observatory (e.g., Land Surface Models, Droughts, Floods in the Mediterranean region).
- Synthesis: Relevant papers are processed by an advanced LLM to generate structured "Extended Cards". These summaries extract key information such as methodology, study configuration, main results, and funding sources.
Note: Whenever possible, we use the full text of the article. however, in some cases, the summary is generated using only the abstract because the full text is not available for automated processing. Some publishers, such as Elsevier or MDPI, are particularly aggressive in blocking automated tools, which can prevent the retrieval of the full paper. - Presentation: The summaries are compiled into this static website for easy access and reading.
In addition to the automated discovery, we manually add articles that are of particular interest to ensure the team remains informed about key developments in our research fields.
Credits
This project is developed and maintained by Pere Quintana Seguí.
Hydrologist and Climate Scientist working on Land Surface Interactions and Climate Extremes.
This project was developed with the assistance of AI tools, specifically Gemini CLI.