『模糊理論』在『中醫辨證』上之研究

中醫辨證論治是中醫治療中最重要的一個核心問題。各種辨證法如病因辯證、臟腑辨證、六經辨證及經絡辨證等。然而並沒有完整一致的評估方法驗證，造成中醫師之間的辨證結果及辨證用語皆有岐異。《中醫證候學》蒐集古今中醫辨證文獻，提出一個整合性的辨證論治體系。本研究即用此書進行症狀標準化，以原書之證候，證型及證類為標準化名詞。然後用資訊工程的模糊邏輯之理論對證門、證類、證型、證候之階層關係定義其模糊權重，實做一資訊化辨證及權重分析系統，以達數位辯證之目的。並與平權變證系統進行比較。探討是否中醫的診斷資料是否可以在一致的系統下，得出滿意的辨證結果。研究結果顯示在證類、證型辨證時，加權和之準確率要比加權比之準確率更高，而在證候辨證時則二者類似。而以書中症狀進行辨證的準確率測試結果發現加權和最低由87.5%至100%；而加權比最低由68.42%至100%；同時平權比則由14.29%至100%。因此由此結果發現若以加權和的方式進行辨證，其結果最少87.5%與《中醫證候學》的內容一致，由此證明中醫可以有一致的辨證結果出現。同時未來中醫界若對辨證的症狀及證型和其相關的症狀結構有一致性的看法的話，此研究可以為未來中醫辨證系統自動化奠定基礎。

關鍵字

中醫自動辨證系統；中醫症狀權重分析；模糊理論

並列摘要

TCM syndrome differentiation is the most critical core issue in TCM treatment. There is no complete and consistent evaluation method for various syndrome differentiation methods, resulting in different syndrome differentiation results among Chinese medicine practitioners. ＂TCM Syndromes＂ to collect ancient and modern TCM dialectical literature, put forward an integrated syndrome differentiation theory system. In this study, the fuzzy logic theory is used to define the relationship between syndromes by weight and to develop the automated TCM syndrome differentiation system, which is compared with the average weight syndrome differentiation system. To explore whether the diagnostic information of Chinese medicine can be consistent and obtain satisfactory results of dialectical. The results of the study showed that the accuracy of syndrome differentiation based on the book symptoms was found in weighted sum method and the lowest accuracy from 87.5% to 100%; while the weighted ratio method was from 68.42% to 100%; and the average weight ratio method from 14.29% to 100%. From this result, we found that at least 87.5% were consistent with the contents of ＂Syndromatology of Traditional Chinese Medicine,＂ if they use weighted sum method, thus proving that TCM syndromes could be consistent. At the same time, if the Chinese medicine practitioners agree symptoms and their related symptoms and structures are consistent, then this study can lay the foundation for the automation of TCM syndrome differentiation system in the future.