Classification of Tea Quality with Fuzzy Cluster Analysis

This paper aims to improve the fuzzy clustering method by proposing the weighted fuzzy c-means clustering method instead. We emphasize on the concept of factor weight combined with the fuzzy clustering analysis, which has been ignored by most of the conventional fuzzy clustering analysis. Weight of factors and membership functions are obtained from experts via sampling survey. The integrated fuzzy classification procedure is developed and fuzzy inference neural networks based on the statistical model is suggested. Finally, as for an empirical example, we apply the proposed technique to the classification for the 69 Taiwan tea qualities. It exhibits that our proposed integrated fuzzy cluster method demonstrates more efficient and better results than traditional ones did.