LASIGE Talks are fortnightly/monthly events to publicize recently distinguished publications or ongoing cutting-edge work by researchers from the research centre, consolidating the scientific culture of the LASIGE community.

Speakers: Matilde Pato and Stefano Stefenon
Date: December 10th, 2025, Wednesday, 12:50
Where: C6.3.27

Program:
12:50 Matilde Pato
12:20 Stefano Stefenon
12:40 Q&A + Break for snacks & coffee

Talk1: Survey on Recommender Systems for Biomedical Items in Life and Health Science – Sur-RS4BioT
Speaker: Matilde Pato
Summary: The generation of biomedical data is of such magnitude that its retrieval and analysis have posed several challenges. A survey of recommender system (RS) approaches in biomedical fields is provided in this analysis, along with a discussion of existing challenges related to large-scale biomedical information retrieval systems. We collect original studies, identify entities and models, and discuss how knowledge graphs (KGs) can improve results. As a result, most of the papers used model-based collaborative filtering algorithms, most of the available datasets did not follow the standard format < user, item, rating >, and regarding qualitative evaluations of RSs use mainly classification metrics. Finally, we have assembled and coded a unique dataset of 60 papers.
Paper: https://doi.org/10.1145/3639047

Talk2:Neural Hierarchical Interpolation Time Series (NHITS) for Reservoir Level Multi Horizon Forecasting in Hydroelectric Power Plants
Speaker: Stefano Stefenon
Summary: Effective reservoir level forecasting is crucial for energy planning and flood prevention in hydroelectric systems. This study addresses the challenge of multi-horizon prediction to support operational decisions and emergency management in dams.
We apply the Neural Hierarchical Interpolation Time Series (NHITS) model to forecast hourly reservoir levels at Brazil’s Barra Grande plant. NHITS achieved an RMSE of 4.64e-4 for one-hour horizons and outperformed benchmarks including LSTM, TCN, and N-BEATS, demonstrating superior accuracy for very short-term predictions.
Paper: https://doi.org/10.1109/ACCESS.2025.3554446