The Application of LSA in TCM Syndromes Classification

Abstract:

The description of syndromes and symptoms in traditional Chinese medicine (TCM) is extremely complicated. And how to diagnose the patient's syndrome in a better way is the primary objective of clinical health care workers all the time. It was a good attempt to diagnose patient's syndrome by combining Latent Semantic Analysis and the feature of TCM knowledge----both syndromes and organs have the same clinical manifestation collection that are symptoms. In this paper, correlative degrees would be computed and sorted in a certain latent semantic space which was constructed by syndromes and organs . According to the result of correlative degrees computing, the classifying could be done by choosing the highest correlative degree as the belonging class. The experimental results show that this method performs quite well.

• Periodical: Applied Mechanics and Materials (Volumes 284-287)
• Pages: 1666-1670
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.284-287.1666
• Online since: January 2013
• Authors: Yan Fang Gan, Zi Wei Ni, Fan Lin
• Keywords: Latent Semantic Analysis (LSA), Singular Value Decomposition (SVD), TCM Diagnosis

References

[1] HE Zhi-lin, WANG Chun-hong. Application of Matrix Singular Value Decomposition (SVD) in Latent Semantic Information Retrieval. Modern Comupter, (2011):21-23.

[2] XuanYungan, ZhuQinghua. Research on Tag Semantic Retrieval in SocialTagging System Based on LSA. Library and Information Service,(2011),55(4)：11-14.

[3] Hu Mingjie,He Yuzhu,Li Jianhong. Fault diagnosis method based on LSA and SVM..IEEE, (2009)：1-4.

DOI: 10.1109/icicisys.2009.5357927