2012 | OriginalPaper | Chapter
Selecting an Appropriate Interestingness Measure to Evaluate the Correlation between Syndrome Elements and Symptoms
Authors : Lei Zhang, Qi-ming Zhang, Yi-guo Wang, Dong-lin Yu
Published in: New Frontiers in Applied Data Mining
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
In order to select the best interestingness measure appropriate for evaluating the correlation between syndrome elements and symptoms, 60 objective interestingness measures were selected from different subjects. Firstly, a hypothesis for a good measure was proposed. Based on the hypothesis, an experiment was designed to evaluate the measures. The experiment was based on the clinical record database of past dynasties including 51,186 clinical cases. The selected dataset in this study had 44,600 records. Han and Re were selected as the experimental syndrome elements. Three indicators calculated according to the distances between two syndrome elements were obtained in the experiment and were combined into one indicator. The
Z score, φ-coefficient
and
Kappa
were selected from 60 measures after the experiment. The
Z score
and
φ- coefficient
were selected according to subjective interestingness. Finally, the
φ- coefficient
was selected as the best measure for its low computational complexity. The method introduced in this paper may be used in other similar territories. Further research of traditional Chinese medicine can be made based on the conclusion made in this paper.