IMPLEMENTASI METODE SUPPORT VECTOR MECHINE DAN K-MEANS CLUSTERING DENGAN SUPPORT VECTOR MECHINE TEMA KARYA ILMIAH PADA STMIK WIDYA CIPTA DHARMA
Abstract
Trends in research topics are useful for building an institutional research roadmap. The classification of themes of scientific works is very much needed to evaluate the vulnerability of research topics of an institution. The implementation of the text mining classification for the theme of scientific works uses the Vector Mechine supprort method and K-means Clustering with Support Vector Mechine (SVM) support. The results of this study are to measure the accuracy of each method used. Support Vector Mechine method at the level of accuracy is 93.33% while the K-means Clustering method with Support Vector Mechine (SVM) accuracy is 99.33%.
References
Gupta, N., “Text Mining for Information Retreival,” Ph.D. thesis, Jaypee Institute of Information Technology
Ian H Witten, Eibe Frank, and Mark A Hall, Data Mining PracticalMachine Learning Tools and Techniques. USA: Elsevier, 2011.
Somantri,O. Wiyono,S. Dairoh., 2014, Metode K-Means Untuk Optimasi Klasifikasi Tema Tugas Akhir Mahasiswa Mengunakan Supprot Vector Machine (SVM). scientific journal of informatic vol 3 no 1 mei 2016 Hal 34-45
Turban, E., Aronson, J.E., Liang, T.P., Introduction to Data Mining, Pearson, 2005.
Yunliang, J., Qing, S., Jing, F., & Xiongtao, Z. 2010,. The Classification for E-government Document Based on SVM. In Web Information Systems and Mining (WISM), 2010 International Conference on (Vol. 2, pp. 257-260).
J.Z. Liang. 2004 “SVM Multi-Classifier And Web Document Classification”, Proceedings of the IEEE Third International Conference on Machine Learning and Cybernetics.
