2026-10-14T14:53:06Z https://www.scidb.cn/oai

10.57760/sciencedb.dmh.01052 2026-05-28T08:42:03Z

2026-05-28 Research on Fine-Grained Intelligent Classification and Grading Techniques for Data in the Traditional Chinese Medicine Industry doi:10.57760/sciencedb.dmh.01052 en In recent years, with the rapid advancement of information technology and the digitalization of Traditional Chinese Medicine (TCM), the TCM industry urgently requires digital transformation. However, this transformation faces critical challenges: semantic fragmentation across industrial links caused by the absence of data standards throughout the entire supply chain, and the inability of general-purpose governance tools to adapt to the strong domain-specific dependency characteristics inherent to TCM. To address these challenges, this thesis proposes and implements a high-granularity intelligent data classification and grading technical system tailored for the TCM industry, achieving deep integration and scenario-based deployment of classification and grading technology across the full TCM industrial chain. First, a three-level data classification and grading management system is constructed, covering the entire industrial chain from “planting—processing—research and development—distribution—application.” Based on data value density, sensitivity, and leakage risk levels, a differentiated control mechanism of “public—restricted—confidential” is established, providing an implementable governance baseline for the industry.Second, a hybrid architecture model integrating domain knowledge graphs and adaptive word segmentation is developed. By constructing a terminology database covering 150,000 geo-authentic medicinal materials and 4,618 natural medicines, the thesis overcomes the adaptation bottleneck of general-purpose segmentation tools in professional terminology boundary splitting.Finally, a closed-loop multimodal intelligent annotation model of “collection—cleaning—pre-annotation—manual review—fine-tuning—application” is designed. The Segment Anything Model (SAM) is integrated to achieve zero-shot image segmentation. The Self-Instruct method is utilized to synthesize low-resource annotation data. Cross-modal knowledge communities are constructed through Graph Retrieval-Augmented Generation (Graph RAG) to support hybrid retrieval and intelligent question answering. In large-scale practice at Yunnan Baiyao Group, this system improved data annotation efficiency by 30 times, enhanced medicinal material quality inspection efficiency by 50%, and saved annual operating costs exceeding ten million yuan. The diagnosis, treatment, and marketing assistant based on Graph RAG has been promoted and applied in thousands of pharmacies. This thesis provides a systematic technical solution and engineering paradigm for the secure circulation and value transformation of data elements in the TCM industry. Traditional Chinese Medicine data; data classification and grading; multimodal data; knowledge graph Mengting.Wu Zhenghan.Xie Xilong.Kang Deli.Meng PUBLIC https://creativecommons.org/licenses/by-nc/4.0/ dataset Science Data Bank