| dc.description.abstract | Metagenome is a study of total DNA from some environmental sources that are directly isolated. The study is conducted by reading the entire DNA of a complete ecosystem (not just one organism). Metagenome refers to the genomic content of complete microbial ecosystems. Since the samples taken from the ecosystems may contain a variety of organisms, it requires a binning process to classify. In this research, k-nearest neighbour (KNN) algorithm was used to classify metagenome fragments and spaced n-mers was used for feature extraction. The research was conducted on two groups of datasets, namely the training organisms and testing organisms with fragment length of 500 bp, 1 kbp, 5 kbp, and 10 kbp. The best accuracy obtained from the training organism dataset reached 99.75% on the fragment test with a length of 10 kbp and k = 3. The highest value of its sensitivity and specificity was also obtained from the same organism dataset, 99.71% and 99.85% respectively. | en |
