Implementasi Minimum Redundancy Maximum Relevance Sebagai Teknik Reduksi Dimensi Pada Klasifikasi Kanker Usus Besar Menggunakan Random Forest

Authors

DOI:

https://doi.org/10.34818/eoe.v7i1.11886

Abstract

Abstrak Kanker merupakan penyakit yang mematikan. Mengutip informasi dari kementrian kesehatan Republik Indonesia pada tahun 2017 sembilan juta orang meninggal akibat kanker. Oleh sebab itu diperlukan sebuah metode untuk mendeteksi kanker salah satunya dengan gen expression. Microarray adalah salah satu teknik dari gen expression. Microarray sendiri memiliki feature yang banyak, feature yang banyak ini tidaklah selalu berkaitan dengan masalah yang sedang dihadapi. Sehingga dibutuhkan teknik reduksi dimensi untuk menyeleksi feature yang bersesuaian dengan masalah yang sedang dihadapi. Pada tugas akhir ini digunakan teknik reduksi dimensi menggunakan Minimum Redundancy Maximum Relevance yang selanjutnya akan disingkat dengan MRMR. Adapun Classifier yang digunakan adalah Random Forest, dimana teknik ini membuat beberapa tree untuk mengklasifikasi data lalu dilakukan voting untuk hasil terbanyak. Persamaan MRMR yang digunakan adalah FCD dan FCQ karena data yang digunakan bernilai kontinu. Setelah semua proses telah dilakukan, diperoleh hasil akurasi dari klasifikasi data microarray dengan menggunakan FCQ sebesar 83,87% dan dengan FCD 61,29%. Kata kunci : microarray, gen expression, random forest, MRMR Abstract Cancer is a deadly disease. Quoting information from the Ministry of Health of the Republic of Indonesia in 2017 nine million people died from cancer. Therefore we need a method to detect cancer, one of which is by gene expression. Microarray is a technique of gene expression. Microarray itself has many features, many of these features are not always related to the problem being faced. So we need a dimension reduction technique to select features that correspond to the problem being faced. In this final project a dimension reduction technique will be used using the Minimum Redundancy Maximum Relevance which will then be abbreviated as MRMR. The Classifier that will be used is Random Forest, where this technique creates several trees to classify data and then will vote for the most results. The MRMR equation used is FCD and FCQ because the data used is continuous. After the process done, the result from classify microarray data using FCQ is 83.87% and with FCD 61.29% Keywords: microarray, gen expression, random forest, MRMR

Downloads

Download data is not yet available.

References

[1] Husna Aydadenta and Adiwijaya. 2018. A Clustering Approach for Feature Selection in Microarray Data Classification Using Random Forest, Journal of Information Processing Systems, 14, 5, (2018), 1167~1175.

[2] Adiwijaya, Kang & Wisesty, Untari Novia & Lisnawati, E. & Aditsania, Annisa & Kusumo, Dana. (2018). Dimensionality Reduction using Principal Component Analysis for Cancer Detection based on Microarray Data Classification. Journal of Computer Science. 14. 1521-1530. 10.3844/jcssp.2018.1521.1530.

[3] Aydadenta, H., & Adiwijaya. 2018, March. On the classification techniques in data mining for microarray data classification. In Journal of Physics: Conference Series (Vol. 971, No. 1, p. 012004). IOP Publishing.

[4] Z. Cao, Y. Wang, Y. Sun, W. Du and Y. Liang, "Effective and stable feature selection method based on filter for gene signature identification in paired microarray data," 2013 IEEE International Conference on Bioinformatics and Biomedicine, Shanghai, 2013, pp. 189-192.

[5] C. Ding and H. Peng, "Minimum redundancy feature selection from microarray gene expression data," Computational Systems Bioinformatics. CSB2003. Proceedings of the 2003 IEEE Bioinformatics Conference. CSB2003, Stanford, CA, USA, 2003, pp. 523-528

[6] Kamus Besar Bahasa Indonesia. Retrieved from https://kbbi.kemdikbud.go.id/entri/kanker.

[7]Kementerian Kesehatan Republik Indonesia. Retrieved from http://www.depkes.go.id/article/print/17020200002/kementerian-kesehatan-ajak-masyarakat-cegah-dan-kendalik an-kanker.html.

[8]Kent Ridge Biomedical Data Set Repository,. (n.d.). Retrieved from http://leo.ugr.es/elvira/DBCRepository/index.html.

[9] D. Pavithra and B. Lakshmanan, "Feature selection and classification in gene expression cancer data," 2017 International Conference on Computational Intelligence in Data Science(ICCIDS), Chennai, 2017, pp. 1-6.

[10] S. Dasgupta, G. Saha, R. Mondal, R. K. Pal and A. Chanda, "A comparison between methods for generating differentially expressed genes from microarray data for prediction of disease," Proceedings of the 2015 Third International Conference on Computer, Communication, Control and Information Technology (C3IT), Hooghly, 2015, pp. 1-5.

[11] C. M. M. Wahid, A. B. M. S. Ali and K. Tickle, "Impact of Feature Selection on Support Vector Machine Using Microarray Gene Expression Data," 2009 Second International Conference on Machine Vision, Dubai, 2009, pp. 189-193.

[12]Ma’ruf, Firda Aminy, Adiwijaya, and Untari Novia Wisesty. "Analysis of the influence of Minimum Redundancy Maximum Relevance as dimensionality reduction method on cancer classification based on microarray data using Support Vector Machine classifier." Journal of Physics: Conference Series. Vol. 1192. No. 1. IOP Publishing, 2019.

[13]Manik, A., Adiwijaya, A., & Utama, D. Q. (2019). Classification of Electrocardiogram Signals using Principal Component Analysis and Levenberg Marquardt Backpropagation for Detection Ventricular Tachyarrhythmia. Journal of Data Science and Its Applications, 2(1), 78-87.

Downloads

Published

2020-04-01

Issue

Pages

2490-4297

Section

Program Studi S1 Informatika

How to Cite

Geramona, I., Adiwijaya, A., & Astuti, W. (2020). Implementasi Minimum Redundancy Maximum Relevance Sebagai Teknik Reduksi Dimensi Pada Klasifikasi Kanker Usus Besar Menggunakan Random Forest. EProceedings of Engineering, 7(1), 2490-4297. https://doi.org/10.34818/eoe.v7i1.11886