Na Li had an accepted paper in 19th e-Science conference


The discovery and reuse of scientific codes are essential in research, with computational notebooks proving effective for sharing such codes. However, locating relevant computational notebooks poses challenges due to their multi-modal nature and lack of evaluation datasets for fair assessments. Previous studies have focused on code-snippet search or content-based approaches, ignoring broader information needs. To tackle these challenges, DeCNR is introduced, employing a fused sparse-dense retrieval model and an evaluation dataset for fair performance evaluation. Experimental results show that DeCNR outperforms baseline methods in terms of F1@5 and NDCG@5, with the system implemented as a web service with REST APIs for seamless integration.

This work has been partially funded by the European Union’s Horizon 2020 research and innovation program by the project CLARIFY under the Marie SkÅ‚odowska-Curie grant agreement No 860627, by the ARTICONF project grant agreement No 825134, by the ENVRI-FAIR project grant agreement No 824068, by the BLUECLOUD project grant agreement No 862409, by the LifeWatch ERIC, by the EPSRC grant EP/W032473/1.