Page 61 - Proceeding The 2nd International Seminar of Science and Technology : Accelerating Sustainable Innovation Towards Society 5.0
P. 61
The 2 International Seminar of Science and Technology
nd
“Accelerating Sustainable innovation towards Society 5.0”
ISST 2022 FST UT 2022
Universitas Terbuka
membership in range 0 to 1, different with logic Boolean which only
have two score namely 0 and 1 [14].
2.3 Fuzzy C-Means
Fuzzy C-Means is method development data grouping from K-Means
where existence each group determined by degree membership
ranges between 0 and 1 [15]. The taller score membership so the taller
degrees membership, so on the contrary the more small score
membership so the more small degrees membership. The following is
a grouping algorithm using the FCM method:
1. Enter the data to be grouped in the form of a matrix of size nxm
(n = number of data, p = attribute of each data).
2. Determine:
a. Number of clusters (c)
b. Weighting power (m)
c. Maximum iterations
d. Smallest error ( ε)
e. Initial objective function (P 0 = 0)
f. Initial iteration (t=1)
3. Generating random numbers as the initial partition matrix (μ ) ,
ik
namely the relative peculiarity matrix of size ixk (i = number of
data, k = number of clusters ) where the row value of each matrix
is 1.
4. Calculate the cluster center ( V )with Equation (1).
kj
w
∑ n i=1 [(U ik ) X ij ]
V = n (1)
kj
∑
i=1 (U ik ) w
5. Calculate the objective function (P) with the following formula:
w
w
P = ∑ n ∑ c k=1 [(∑ m (X -V ) )(U ) ]
kj
j=1
ij
ik
i=1
6. Fixed the relative peculiarity matrix ( μ ) in Equation (2).
ik
-1
2
[∑ m ((X ij -V kj ) )] w-1
j=1
μ = -1 (2)
ik
2
∑ n [∑ m ((X ij -V kj ) )] w-1
i=1 j=1
40 ISST 2022 – FST Universitas Terbuka, Indonesia
International Seminar of Science and Technology “Accelerating Sustainable
Towards Society 5.0