Novelty Scores for "Triadic Novelty: A Typology and Measurement Framework for Recognizing Novel Contributions in Science"

This repository contains the derived novelty scores used in the study titled “Triadic Novelty: A Typology and Measurement Framework for Recognizing Novel Contributions in Science.” Each score is linked to individual records from the Web of Science Raw Data (2022 version) via their unique Web of Science IDs. These novelty scores were calculated using the Triadic Novelty mesaures proposed in the paper and correspond to the dataset employed in the regression analyses. Specifically, the Triadic Novelty measures and their normalized versions are available through an open-source Python package named "triadic-novelty." Full methodological details are provided in the associated preprint: https://doi.org/10.48550/arXiv.2506.17851. We share this dataset to support transparency, reproducibility, and further exploration of novelty in scientific research. Important note on data sharing: The Web of Science Raw Data (2022 Version) employed in this paper was obtained from the Web of Science under a specific institutional agreement, which forbids the authors from sharing data derivatives. Therefore, this repository includes only the derived novelty scores and associated WoS identifiers, not the full metadata or other bibliographic records.