Mengenal Machine Learning Dengan Teknik Supervised Dan Unsupervised Learning Menggunakan Python

  • Endang Retnoningsih Sistem Informasi; Universitas Bina Insani
  • Rully Pramudita Manajemen Informatika; Universitas Bina Insani

Abstract

Abstrak: Machine learning merupakan sistem yang mampu belajar sendiri untuk memutuskan sesuatu tanpa harus berulangkali diprogram oleh manusia sehingga komputer menjadi semakin cerdas berlajar dari pengalaman data yang dimiliki. Berdasarkan teknik pembelajarannya, dapat dibedakan supervised learning menggunakan dataset (data training) yang sudah berlabel, sedangkan unsupervised learning menarik kesimpulan berdasarkan dataset. Input berupa dataset digunakan pembelajaran mesin untuk menghasilkan analisis yang benar. Permasalahan yang akan diselesaikan bunga iris (iris tectorum) yang memiliki bunga bermaca-macam warna dan memiliki sepal dan petal yang menunjukkan spesies bunga, dibutuhkan metode yang tepat untuk pengelompokan bunga-bunga tersebut kedalam spesiesnya iris-setosa, iris-versicolor atau iris-virginica. Penyelesaian digunakan Python yang menyediakan algoritma dan library yang digunakan membuat machine learning. Penyelesaian dengan teknik supervised learning dipilih algoritma KNN Clasiffier dan teknik unsupervised learning dipilih algoritma DBSCAN Clustering. Hasil yang diperoleh Python menyediakan library yang lengkap numPy, Pandas, matplotlib, sklearn untuk membuat pemrograman machine learning dengan algortima KNN memanggil from sklearn import neighbors termasuk teknik supervised, maupun DBSCAN memanggil from sklearn.cluster import DBSCAN termasuk teknik unsupervised learning. Kemampuan Python memberikan hasil output sesuai input dalam dataset menghasilkan keputusan berupa klasifikasi maupun klusterisasi.
 
Kata kunci: DBSCAN, KNN, machine learning, python.
 
Abstract: Machine learning is a system that is able to learn on its own to decide something without having to be repeatedly programmed by humans so that computers become smarter in learning from the experience of the data they have. Based on the learning technique, supervised learning can be distinguished using a dataset (training data) that is already labeled, while unsupervised learning draws conclusions based on the dataset. The input in the form of a dataset is used by machine learning to produce the correct analysis. The problem to be solved by iris flowers (iris tectorum), which has flowers of various colors and has sepals and petals that indicate the species of flowers, requires an appropriate method for grouping these flowers into iris-setosa, iris-versicolor or iris-virginica species. The solution is used by Python, which provides the algorithms and libraries used to make machine learning. The solution with the supervised learning technique was chosen by the KNN Clasiffier algorithm and the unsupervised learning technique was selected by the DBSCAN Clustering algorithm. The results obtained by Python provide a complete library of numPy, Pandas, matplotlib, sklearn to create machine learning programming with KNN algorithms calling from sklearn import neighbors including supervised techniques, and DBSCAN calling from sklearn.cluster import DBSCAN including unsupervised learning techniques. Python's ability to provide output according to the input in the dataset results in decisions in the form of classification and clustering.
 
Keywords: DBSCAN, KNN, machine learning, python.

References

[1] P. D. Kusuma, Machine Learning Teori, Program, dan Studi Kasus. Yogyakarta: Deepublish, 2020.
[2] Z. A. Fikriya, M. I. Irawan, and Soetrisno, “Implementasi Extreme Learning Machine untuk Pengenalan Objek Citra Digital,” J. Sains dan Seni ITS, vol. 6, no. 1, pp. A18–A23, 2017.
[3] Nurhayati, Busman, and R. P. Iswara, “Pengembangan Algoritma Unsupervised Learning Technique Pada Big Data Analysis di Media Sosial sebagai media promosi Online Bagi Masyarakat,” J. Tek. Inform., vol. 12, no. 1, pp. 79–96, 2019.
[4] F. S. Pamungkas, B. D. Prasetya, and I. Kharisudin, “Perbandingan Metode Klasifikasi Supervised Learning pada Data Bank Customers Menggunakan Python,” in Prosiding Seminar Nasional Matematika, 2020, vol. 3, pp. 689–694.
[5] F. A. Bachtiar, I. K. Syahputra, and S. A. Wicaksono, “Perbandingan Algoritme Machine Learning untuk Memprediksi Pengambil Matakuliah,” J. Teknol. Inf. dan Ilmu Komput., vol. 6, no. 5, p. 543, 2019.
[6] B. S. Ashari, S. C. Otniel, and Rianto, “Perbandingan Kinerja K-Means Dengan DSCAN Untuk Metode Clustering Data Penjualan Online Retail,” J. Siliwangi, vol. 5, no. 2, pp. 72–77, 2019.
[7] I. Muhammad and Z. Yan, “Supervised Machine Learning Approaches: a Survey,” ICTACT J. Soft Comput., vol. 05, no. 03, pp. 946–952, 2015.
[8] Fahrizal, F. O. Reynaldi, and N. Hikmah, “Implementasi Machine Learning pada Siatem PETS Identification Menggunakan Python Berbasis UBuntu,” J. Inf. Syst. Informatics Comput., vol. 4, no. 1, pp. 86–91, 2020.
[9] H. Herlawati and R. T. Handayanto, “Penggunaan Matlab dan Python dalam Klasterisasi Data,” J. Kaji. Ilm., vol. 20, no. 1, pp. 103–118, 2020.
[10] T. Wahyono, Fundamental of Python for Machine Learning: Dasar-Dasar Pemrograman Python untuk Machine Learning dan Kecerdasan Buatan. Yogyakaerta: Gava Media, 2018.
[11] http://www.python.org/, “Python.” [Online]. Available: http://www.python.org/. [Accessed: 11-Nov-2020].
[12] http://repo.anaconda.com/, “Anaconda.” [Online]. Available: http://repo.anaconda.com/. [Accessed: 11-Nov-2020].
[13] A. Wanto et al., Data Mining : Algoritma dan Implementasi. Medan: Yayasan Kita Menulis, 2020.
[14] http://archive.ics.uci.edu/ml/datasets/Iris, “UCI Machine Learning Repository: Iris Data Set.” [Online]. Available: http://archive.ics.uci.edu/ml/datasets/Iris. [Accessed: 11-Nov-2020].
Published
2020-12-28
How to Cite
RETNONINGSIH, Endang; PRAMUDITA, Rully. Mengenal Machine Learning Dengan Teknik Supervised Dan Unsupervised Learning Menggunakan Python. BINA INSANI ICT JOURNAL, [S.l.], v. 7, n. 2, p. 156-165, dec. 2020. ISSN 2527-9777. Available at: <http://ejournal-binainsani.ac.id/index.php/BIICT/article/view/1422>. Date accessed: 18 apr. 2021.