Optimasi Parameter Pada Metode Support Vector Machine Berbasis Algoritma Genetika untuk Estimasi Kebakaran Hutan

Hani Harafani; Romi Satria Wahono

Optimasi Parameter Pada Metode Support Vector Machine Berbasis Algoritma Genetika untuk Estimasi Kebakaran Hutan

Hani Harafani, Romi Satria Wahono

Abstract

Kebakaran hutan merupakan salah satu masalah lingkungan yang mengancam hutan, menimbulkan dampak negatif pada lingkungan, menciptakan masalah ekonomi, dan kerusakan ekologis, serta menyebabkan kerugian penting di seluruh dunia setiap tahunnya. Estimasi area yang terbakar penting dilakukan, karena area yang terbakar dapat mencerminkan berapa kuat radiasi api pada vegetasi disekitarnya. SVM dapat mengatasi masalah klasifikasi dan regresi linier ataupun nonlinier kernel yang dapat menjadi satu kemampuan algoritma pembelajaran untuk klasifikasi serta regresi. Namun, SVM juga memiliki kelemahan yaitu sulitnya menentukan nilai parameter yang optimal. Untuk menyelesaikan permasalahan tersebut algoritma genetika diusulkan untuk diterapkan sebagai algoritma pencarian nilai parameter yang efisien pada SVM. Beberapa eksperimen dilakukan untuk menghasilkan estimasi yang akurat. Awalnya percobaan dilakukan pada kernel –kernel SVM (dot, RBF, polynomial) untuk menentukan kernel mana yang akan digunakan, kemudian model SVM+GA juga dibandingkan dengan model regresi lainnya seperti Linear Regression, k-NN, dan Neural Network. Berdasarkan eksperimen dengan 10 kombinasi parameter pada metode SVM dan SVM+GA dengan kernel dot, RMSE terkecil dihasilkan oleh model SVM+GA sebesar 1.379, sementara pada percobaan SVM dan SVM+GA dengan kernel polynomial RMSE terkecil diperoleh model SVM+GA sebesar 1.379, sedangkan pada percobaan SVM dan SVM+GA dengan kernel RBF diperoleh RMSE terkecil pada model SVM+GA sebesar 1.379.Selanjutnya berdasarkan perbandingan rata-rata RMSE, kernel RBF unggul dengan nilai RMSE terkecil yaitu 1.432 pada SVM, dan 1.418 pada SVM+GA. Pada perbandingan nilai rata-rata RMSE antara SVM(RBF)+GA dengan model lainnya, RMSE terkecil dihasilkan oleh SVM(RBF)+GA yaitu sebesar 1.418, disusul dengan model SVM(RBF) sebesar 1.432, keudian Linear Regression sebesar 1.459, dilanjutkan oleh model k-NN sebesar 1.526 dan yang terakhir adalah NN dengan nilai RMSE sebesar 1.559. maka dapat disimpulkan bahwa optimasi parameter yang dilakukan GA pada model SVM terbukti dapat mengurangi tingkat error pada model SVM tanpa optimasi parameter pada dataset forestfire, selain model SVM(RBF)+GA pada penelitian ini juga terbukti lebih baik dari model regresi lainnya

Full Text:

PDF

References

Brun, C., Margalef, T., & Cortés, A. (2013). Coupling Diagnostic and Prognostic Models to a Dynamic Data Driven Forest Fire Spread Prediction System. Procedia Computer Science, 18, 1851–1860. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S1877050913004973

Chen, K.-Y. (2007). Forecasting systems reliability based on support vector regression with genetic algorithms. Reliability Engineering & System Safety, 92(4), 423–432.

Conway, D., & White, J. M. (2012). Machine Learning for Hackers. (J. Steele, Ed.).

Cortez, P., & Morais, A. (2007). A Data Mining Approach to Predict Forest Fires using Meteorological Data.

Denham, M., Wendt, K., Bianchini, G., Cortés, A., & Margalef, T. (2012). Dynamic Data-Driven Genetic Algorithm for forest fire spread prediction. Journal of Computational Science, 3(5), 398–404. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S1877750312000658

Dua. (2011). Data Mining and Machine Learning in Cybersecurity. (Dua, Ed.).

Eastaugh, C. S., & Hasenauer, H. (2014). Deriving forest fire ignition risk with biogeochemical process modelling. Environmental Modelling & Software, 55, 132–142. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S1364815214000280

Gorunescu, F. (2011). Intelligent Systems Reference Library. (Gorunescu, Ed.).

Gu, J., Zhu, M., & Jiang, L. (2011). Housing price forecasting based on genetic algorithm and support vector machine. Expert Systems with Applications, 38(4), 3383–3386. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0957417410009310

Guo, X., Li, D., & Zhang, A. (2012). Improved Support Vector Machine Oil Price Forecast Model Based on Genetic Algorithm Optimization Parameters. AASRI Procedia, 1, 525–530. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S2212671612000832

Hosseini, M., Javaherian, A., & Movahed, B. (2014). Determination of permeability index using Stoneley slowness analysis, NMR models, and formation evaluations: a case study from a gas reservoir, south of Iran. Journal of Applied Geophysics, 109, 80–87.

Ilhan, I., & Tezel, G. (2013). A genetic algorithm-support vector machine method with parameter optimization for selecting the tag SNPs. Journal of Biomedical Informatics, 46(2), 328–40. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/23262450

Jia, Z., Ma, J., Wang, F., & Liu, W. (2011). Hybrid of simulated annealing and SVM for hydraulic valve characteristics prediction. Expert Systems with Applications, 38(7), 8030–8036.

Jia, Z.-Y., Ma, J.-W., Wang, F.-J., & Liu, W. (2010). Characteristics forecasting of hydraulic valve based on grey correlation and ANFIS. Expert Systems with Applications, 37(2), 1250–1255. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0957417409005624

Kaytez, F., Taplamacioglu, M. C., Cam, E., & Hardalac, F. (2015).

Electrical Power and Energy Systems Forecasting electricity consumption : A comparison of regression analysis , neural networks and least squares support vector machines. International Journal of Electrical Power and Energy Systems, 67, 431–438.

Lee, S., Kang, P., & Cho, S. (2014). Neurocomputing Probabilistic local reconstruction for k -NN regression and its application to virtual metrology in semiconductor manufacturing. Neurocomputing, 131, 427–439.

Lira, M. A. T., Da Silva, E. M., Alves, J. M. B., & Veras, G. V. O. (2014). Estimation of wind resources in the coast of Ceará, Brazil, using the linear regression theory. Renewable and Sustainable Energy Reviews, 39, 509–529.

Machairas, V., Tsangrassoulis, A., & Axarli, K. (2014). Algorithms for optimization of building design: A review. Renewable and Sustainable Energy Reviews, 31(1364), 101–112. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S1364032113007855

Maimon, O., & Rokach, L. (2010). Data Mining and Knowledge Discovery Handbook.

Özbayoğlu, a. M., & Bozer, R. (2012). Estimation of the Burned Area in Forest Fires Using Computational Intelligence Techniques. Procedia Computer Science, 12, 282–287.

Pan, S., Iplikci, S., Warwick, K., & Aziz, T. Z. (2012). Parkinson’s Disease tremor classification – A comparison between Support Vector Machines and neural networks. Expert Systems with Applications, 39(12), 10764–10771.

Quintano, C., Fernández-Manso, A., Stein, A., & Bijker, W. (2011). Estimation of area burned by forest fires in Mediterranean countries: A remote sensing data mining perspective. Forest Ecology and Management, 262(8), 1597–1607. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S0378112711004385

Raghavendra. N, S., & Deka, P. C. (2014). Support vector machine applications in the field of hydrology: A review. Applied Soft Computing, 19, 372–386. Retrieved from http://linkinghub.elsevier.com/retrieve/pii/S1568494614000611

Rynkiewicz, J. (2012). General bound of overfitting for MLP regression models. Neurocomputing, 90, 106–110.

Singh, P., & Borah, B. (2014). International Journal of Approximate Reasoning Forecasting stock index price based on M-factors fuzzy time series and particle swarm optimization. International Journal of Approximate Reasoning, 55(3), 812–833.

Suganyadevi, M. V, & Babulal, C. K. (2014). Support Vector Regression Model for the prediction of Loadability Margin of a Power System. Applied Soft Computing Journal, 24, 304–315.

Tiryaki, S., Öz, Ş., & Y, İ. (2014). International Journal of Adhesion & Adhesives Comparison of arti fi cial neural network and multiple linear regression models to predict optimum bonding strength of heat treated woods, 55, 29–36.

Wang, X., Wen, J., Zhang, Y., & Wang, Y. (2014). Optik Real estate price forecasting based on SVM optimized by PSO. Optik - International Journal for Light and Electron Optics, 125(3), 1439–1443.

Yang, X. (2014). Nature-Inspired Optimization Algorithms. Elsevier. doi:10.1016/B978-0-12-416743-8.00005-1

Yilmaz, I., & Kaynar, O. (2011). Multiple regression, ANN (RBF, MLP) and ANFIS models for prediction of swell potential of clayey soils. Expert Systems with Applications, 38(5), 5958–5966.

Zameer, A., Mirza, S. M., & Mirza, N. M. (2014). Core loading pattern optimization of a typical two-loop 300MWe PWR using Simulated Annealing (SA), novel crossover Genetic Algorithms (GA) and hybrid GA(SA) schemes. Annals of Nuclear Energy, 65, 122–131.

Zhang, D., Liu, W., Wang, A., & Deng, Q. (2011). Parameter Optimization for Support Vector Regression Based on Genetic Algorithm with Simplex Crossover Operator. Journal of Information & Computational Science, 6(June), 911–920. Retrieved from http://www.joics.com/publishedpapers/2011_8_6_911_920.pdf

Zhao, M., Fu, C., Ji, L., Tang, K., & Zhou, M. (2011). Feature selection and parameter optimization for support vector machines: A new approach based on genetic algorithm with feature chromosomes. Expert Systems with Applications, 38(5), 5197–5204.

Zhao, W., Tao, T., & Zio, E. (2015). System reliability prediction by support vector regression with analytic selection and genetic algorithm parameters selection. Applied Soft Computing, 30, 792–802.

Refbacks

There are currently no refbacks.