Page 62 - Proceeding The 2nd International Seminar of Science and Technology : Accelerating Sustainable Innovation Towards Society 5.0
P. 62
nd
The 2 International Seminar of Science and Technology
“Accelerating Sustainable innovation towards Society 5.0”
ISST 2022 FST UT 2022
Universitas Terbuka
7. Check condition stop:
a. If | Pt – Pt-1 | < ε or t > Maxiter then stop
b. If no : t = t+1 then repeat step 4
2.4 Fuzzy Possibilistic C-Means
Fuzzy Possibilistic C - Means (FPCM) is something algorithm
development from Fuzzy C- Means (FCM) and Possibilistic C Means
(PCM) [16]. Step- step analysis on the FPCM method, namely as
following:
1. Enter the data (Xij) to be grouped in the form of a matrix of size
nxm (n = number of data, p = number of variables).
2. Determine:
a. Number of clusters (c ≥ 2)
b. FCM weighting power (w > 1)
c. PCM weighting power (η > 1
d. Maximum iterations
e. Smallest error ( ε)
f. Initial objective function (P 0 = 0)
g. Initial iteration (t=1)
3. Call matrix peculiarity relatively (μ ) and the cluster center (V kj )
ik
in the final result FCM method for count matrix peculiarity
absolute which is in the form of a relative peculiarity matrix ( t ik )
with Equation (3).
-1
2
[∑ m ((X ij -V kj ) )] η-1
tik = j=1 -1 (3)
2
∑ n i=1 [∑ m ((X ij -V kj ) )] η-1
j=1
4. Fixed cluster center ( V ) with Equation (4).
kj
w
∑ n (μ ik +t ik η )X ij
V = i=1 w η (4)
kj
n
∑
i=1 (μ ik +t ik )
5. Fix μ with Equation (5).
ik
-1
2
[∑ m ((X ij -V kj ) )] w-1
μ = j=1 -1 (5)
ik
2
∑ n [∑ m ((X ij -V kj ) )] w-1
i=1 j=1
ISST 2022 – FST Universitas Terbuka, Indonesia 41
International Seminar of Science and Technology “Accelerating Sustainable
Towards Society 5.0