An Improved Fuzzy C-Means Clustering Algorithm and Application in Meteorological Data
The fuzzy clustering algorithm is sensitive to the m value and the degree of membership. Because of the deficiencies of traditional FCM clustering algorithm, we made specific improvement. Through the calculation of the value of m, the amendments of degree of membership to the discussion of issues, effectively compensate for the deficiencies of the traditional algorithm and achieve a relatively good clustering effect. Finally, through the analysis of temperature observation data of the three northeastern province of china in 2000, the reasonableness of the method is verified.
Qi Luo and Yuanzhi Wang
H. F. Li et al., "An Improved Fuzzy C-Means Clustering Algorithm and Application in Meteorological Data", Advanced Materials Research, Vols. 181-182, pp. 545-550, 2011