[Methodology study of classification algorithm in traditional Chinese medicine syndrome study]

Area of Science:

• Computational intelligence
• Traditional Chinese Medicine (TCM)
• Syndrome classification

Context:

• Modernizing Traditional Chinese Medicine (TCM) requires robust classification and diagnostic criteria for its syndromes.
• Data mining offers powerful tools for analyzing complex TCM syndrome data.
• Understanding algorithm suitability is key to advancing TCM research.

Purpose:

• To review and analyze the application of data mining classification algorithms in Traditional Chinese Medicine (TCM) syndrome research.
• To identify the features and suitability of various algorithms for different TCM research objectives.
• To explore the potential of hybrid algorithms for improved TCM syndrome classification.

Summary:

• The study examines classification algorithms such as rough sets, cluster analysis, fuzzy sets, neural networks, and decision trees for TCM syndrome research.
• Rough sets and cluster analysis are suited for exploratory research, while fuzzy sets, neural networks, and decision trees are better for clear diagnostic criteria.
• Hybrid approaches like fuzzy clustering and fuzzy rough sets show promise for TCM syndrome classification.

Impact:

• Provides guidance on selecting appropriate data mining algorithms for TCM syndrome research.
• Highlights the need for developing novel algorithms tailored to the unique characteristics of TCM syndromes.
• Facilitates the modernization and standardization of TCM through advanced computational methods.