Modifikasi Algoritme J-Bit Encoding untuk Meningkatkan Rasio Kompresi
Abstract: J-bit encoding is a
lossless data compression algorithm that manipulates each bit of file data in
order to minimize the size by dividing data into two outputs and combining data
into two outputs. This research proposes a modification of the J-bit Encoding algorithm
by eliminating zero and one symbols of the first output. As a result, the first
output will contain the original data without zero and one symbols and the
second output will contain the value of two bits that describe the position of
zero, one, and byte besides zero and one. The two algorithms are compared by
testing four scheme combination algorithms, which are (i) Burrows-Wheeler
transformation, Move to Front, J-Bit Encoding, and arithmetic coding, (ii)
Burrows-Wheeler transformation, Move to Front, modification of the J-bit
Encoding, and arithmetic coding, (iii) Burrows-Wheeler transformation, Move One
From Front, J-Bit encoding, and arithmetic coding, (iv) Burrows-Wheeler
transformation, Move One From Front, modification of the J-bit Encoding, and
arithmetic coding. By using the Calgary Corpus and Canterbury Corpus data sets,
the test results show that the best compression ratio is obtained by using a
second scheme on average. On the other hand, by using four image files, the
test results show that the best compression ratio is obtained by using a fourth
scheme on average.
Kata Kunci: kompresi data, burrows-wheeler compressio algorithm, j-bit
encoding, kombinasi algoritme
Penulis: Johanes K.M. Lobang,
Pranowo, Suyoto
Kode Jurnal: jptlisetrodd170180
