Rather than improving algorithmic c

Rather than improving algorithmic complexity of a specific
clustering method, data abstraction aims to scale down
a large data set with minimum loss of information for efficient
clustering. The large literature on data abstraction
includes (but is not limited to) such widely used methods as:
random sampling (e.g., CLARANS [31]); selection of representative
points (e.g., CURE [19], data bubble [9]); usage of
cluster prototypes(e.g., Stream [18]) and sufficient statistics
(e.g., BIRCH [41], scalable k-means [8], CluStream [3], data
squashing [12]); grid-based quantization [5, 20] and sparcification
of connectivity or distance matrix (e.g., CHAMELEON
[24]).

Results (Indonesian) 2:

Daripada meningkatkan kompleksitas algoritmik dari spesifik
metode clustering, abstraksi data bertujuan untuk menurunkan
data yang besar dengan kehilangan minimal informasi yang efisien
clustering. Literatur besar pada abstraksi data yang
meliputi (namun tidak terbatas pada) seperti metode banyak digunakan sebagai:
random sampling (misalnya, CLARANS [31]); pemilihan perwakilan
poin (misalnya, CURE [19], Data bubble [9]); penggunaan
prototipe klaster (misalnya, Streaming [18]) dan cukup statistik
(misalnya, BIRCH [41], scalable k-means [8], CluStream [3], data yang
meremas [12]); grid berbasis kuantisasi [5, 20] dan sparcification
konektivitas atau jarak matriks (misalnya, CHAMELEON
[24]).