The National Data Administration convened a symposium on the pilot and pioneering work of data annotation to advance the construction of high-quality datasets through "supply determined by demand" and "construction promoted by application."

According to the official WeChat account of the National Data Administration, the National Data Administration organized a meeting in Beijing on the afternoon of October 27 to advance the pilot work on data annotation, summarize the progress of the pilot work on data annotation, exchange typical experiences, and deploy key tasks for the next stage. The meeting emphasized that data annotation has the notable characteristic of being scenario-based and requires coordinated efforts and joint planning by all localities and entities. For the next step, first, it is necessary to "match supply to demand" and form a demand map for datasets, conduct a statistical survey of upstream and downstream entities associated with the pilot cities for data annotation, and understand the demand for high-quality datasets in the near term and medium-to-long term. Second, it is necessary to "promote construction through application" and advance the implementation of "model-data resonance" scenarios, guiding the pilot cities for data annotation, the chain leaders in the construction of industry high-quality datasets, and the pilot units to test the effectiveness of their work through scenario application results. Third, it is necessary to continue guiding annotated datasets to be registered in the national dataset management service system, track the application of datasets that have passed quality evaluation, and promote the effective reuse and expansion of datasets.