Subjects: Library Science，Information Science >> Information Science submitted time 2023-04-14
Abstract： Purpose/Significance Combining scientific research team with algorithmic innovation and exploring its influence on algorithmic innovation from the perspective of scientific research team traits can help improve the academic capacity of scientific research team, which in turn can promote algorithmic innovation and scientific innovation. Method/Process Three research team measures, namely, research team size, number of research team institutions, and research team institution type, and two algorithm innovation evaluation indexes, namely, algorithm performance and academic output, are established. Taking 543 scientific research teams under the image classification task in the field of machine learning as an example, non-parametric tests and multiple linear regression models were used to explore the effect of scientific research team traits on algorithm innovation, and suggestions to enhance the scientific research performance of scientific research teams and promote algorithm reinvention were proposed based on the empirical results. Result/Conclusion All of the research team characteristics measures have an effect on algorithm innovation, mainly: the type of research team institution has a significant effect on algorithm innovation, the hybrid research team has the best performance in algorithm model accuracy, and the enterprise research team has the best performance in algorithm paper citations; the research team size has a positive effect on algorithm performance and academic output; the research team The number of institutions has a positive effect on the academic output and a negative effect on the algorithm performance.
Peer Review Status:
Awaiting Review
Hits
Hits
Downloads
Comment