Support Vector Machine Berbasis Feature Selection Untuk Sentiment Analysis Kepuasan Pelanggan Terhadap Pelayanan Warung dan Restoran Kuliner Kota Tegal

Penulis

  • Oman Somantri Politeknik Harapan Bersama Tegal
  • Dyah Apriliani Politeknik Harapan Bersama Tegal

DOI:

https://doi.org/10.25126/jtiik.201855867

Kata Kunci:

Sentiment Analysis, Support Vector Machine (SVM), feature selection, Information Gain (IG), Chi Square

Abstrak

Abstrak

 

Setiap pelanggan pasti menginginkan sebuah pendukung keputusan dalam menentukan pilihan ketika akan mengunjungi sebuah tempat makan atau kuliner yang sesuai dengan keinginan salah satu contohnya yaitu di Kota Tegal. Sentiment analysis digunakan untuk memberikan sebuah solusi terkait dengan permasalahan tersebut, dengan menereapkan model algoritma Support Vector Machine (SVM). Tujuan dari penelitian ini adalah mengoptimalisasi model yang dihasilkan dengan diterapkannya feature selection menggunakan algoritma Informatioan Gain (IG) dan Chi Square pada hasil model terbaik yang dihasilkan oleh SVM pada klasifikasi tingkat kepuasan pelanggan terhadap warung dan restoran kuliner di Kota Tegal sehingga terjadi peningkatan akurasi dari model yang dihasilkan. Hasil penelitian menunjukan bahwa tingkat akurasi terbaik dihasilkan oleh model SVM-IG dengan tingkat akurasi terbaik sebesar 72,45% mengalami peningkatan sekitar 3,08% yang awalnya 69.36%. Selisih rata-rata yang dihasilkan setelah dilakukannya optimasi SVM dengan feature selection adalah 2,51% kenaikan tingkat akurasinya. Berdasarkan hasil penelitian bahwa feature selection dengan menggunakan Information Gain (IG) (SVM-IG) memiliki tingkat akurasi lebih baik apabila dibandingkan SVM dan Chi Squared (SVM-CS) sehingga dengan demikian model yang diusulkan dapat meningkatkan tingkat akurasi yang dihasilkan oleh SVM menjadi lebih baik.


Abstract

 

The Customer needs to get a decision support in determining a choice when they’re visit a culinary restaurant accordance to their wishes especially at Tegal City. Sentiment analysis is used to provide a solution related to this problem by applying the Support Vector Machine (SVM) algorithm model. The purpose of this research is to optimize the generated model by applying feature selection using Informatioan Gain (IG) and Chi Square algorithm on the best model produced by SVM on the classification of customer satisfaction level based on culinary restaurants at Tegal City so that there is an increasing accuracy from the model. The results showed that the best accuracy level produced by the SVM-IG model with the best accuracy of 72.45% experienced an increase of about 3.08% which was initially 69.36%. The difference average produced after SVM optimization with feature selection is 2.51% increase in accuracy. Based on the results of the research, the feature selection using Information Gain (SVM-IG) has a better accuracy rate than SVM and Chi Squared (SVM-CS) so that the proposed model can improve the accuracy of SVM better.

Downloads

Download data is not yet available.

Referensi

AMINUDIN, A., SN, A. and AHMAD, B., 2018. Automatic Question Generation (AQG) Dari Dokumen Teks Bahasa Indonesia Berdasarkan Non-Factoid Quesion. Jurnal Teknologi Informasi dan Ilmu Komputer, [online] 5(2), p.217. Available at: <http://jtiik.ub.ac.id/index.php/jtiik/article/view/664>.

DI CARO, L. and GRELLA, M., 2013. Sentiment analysis via dependency parsing. Computer Standards and Interfaces, [online] 35(5), pp.442–453. Available at: <http://dx.doi.org/10.1016/j.csi.2012.10.005>.

CHEN, J., HUANG, H., TIAN, S. and QU, Y., 2009. Feature selection for text classification with Na??ve Bayes. Expert Systems with Applications, [online] 36(3 PART 1), pp.5432–5435. Available at: <http://dx.doi.org/10.1016/j.eswa.2008.06.054>.

CHEN, K., ZHANG, Z., LONG, J. and ZHANG, H., 2016. Turning from TF-IDF to TF-IGM for term weighting in text classification. Expert Systems with Applications, [online] 66, pp.1339–1351. Available at: <http://dx.doi.org/10.1016/j.eswa.2016.09.009>.

FRIEDRICHS, F. and IGEL, C., 2005. Evolutionary tuning of multiple SVM parameters. Neurocomputing, [online] 64, pp.107–117. Available at: <http://linkinghub.elsevier.com/retrieve/pii/S0925231204005223>.

VAN DER GAAG, M., HOFFMAN, T., REMIJSEN, M., HIJMAN, R., DE HAAN, L., VAN MEIJEL, B., VAN HARTEN, P.N., VALMAGGIA, L., DE HERT, M., CUIJPERS, A. and WIERSMA, D., 2006. The five-factor model of the Positive and Negative Syndrome Scale II: A ten-fold cross-validation of a revised model. Schizophrenia Research, 85(1–3), pp.280–287.

JIU-ZHEN LIANG, 2004. SVM multi-classifier and Web document classification. In: Proceedings of 2004 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.04EX826). [online] IEEE, pp.1347–1351. Available at: <http://ieeexplore.ieee.org/document/1381982/>.

KANG, H., YOO, S.J. and HAN, D., 2012. Senti-lexicon and improved Naïve Bayes algorithms for sentiment analysis of restaurant reviews. Expert Systems with Applications, [online] 39(5), pp.6000–6010. Available at: <http://dx.doi.org/10.1016/j.eswa.2011.11.107>.

KONCZ, P. and PARALIC, J., 2011. An approach to feature selection for sentiment analysis. In: 2011 15th IEEE International Conference on Intelligent Engineering Systems. [online] IEEE, pp.357–362. Available at: <http://ieeexplore.ieee.org/document/5954773/>.

LIU, B., 2010. Sentiment Analysis and Subjectivity. 2nd ed. [online] Handbook of natural language processing. Available at: <ftp://nozdr.ru/biblio/kolxo3/Cs/CsNl/Indurkhya N., Damerau F.J. (eds.) Handbook of natural language processing (2ed., CRC, 2010)(ISBN 9781420085921)(O)(692s)_CsNl_.pdf#page=653>.

MEDHAT, W., HASSAN, A. and KORASHY, H., 2014. Sentiment analysis algorithms and applications: A survey. Ain Shams Engineering Journal, [online] 5(4), pp.1093–1113. Available at: <http://dx.doi.org/10.1016/j.asej.2014.04.011>.

MORAES, R., VALIATI, J.F. and GAVIÃO NETO, W.P., 2013. Document-level sentiment classification: An empirical comparison between SVM and ANN. Expert Systems with Applications, [online] 40(2), pp.621–633. Available at: <http://dx.doi.org/10.1016/j.eswa.2012.07.059>.

RAVI, K. and RAVI, V., 2015. A survey on opinion mining and sentiment analysis: Tasks, approaches and applications. [online] Knowledge-Based Systems, Elsevier B.V. Available at: <http://dx.doi.org/10.1016/j.knosys.2015.06.015>.

REYES, A. and ROSSO, P., 2012. Making objective decisions from subjective data: Detecting irony in customer reviews. Decision Support Systems, [online] 53(4), pp.754–760. Available at: <http://dx.doi.org/10.1016/j.dss.2012.05.027>.

ROBALDO, L. and DI CARO, L., 2013. OpinionMining-ML. Computer Standards and Interfaces, 35(5), pp.454–469.

SOMANTRI, O. and KHAMBALI, M., 2017. Feature Selection Klasifikasi Kategori Cerita Pendek Menggunakan Naïve Bayes dan Algoritme Genetika. Jurnal Nasional Teknik Elektro dan Teknologi Informasi (JNTETI), [online] 6(3). Available at: <http://ejnteti.jteti.ugm.ac.id/index.php/JNTETI/article/view/332/257> [Accessed 13 Oct. 2017].

TAN, S. and ZHANG, J., 2008. An empirical study of sentiment analysis for chinese documents. Expert Systems with Applications, [online] 34(4), pp.2622–2629. Available at: <http://linkinghub.elsevier.com/retrieve/pii/S0957417407001534>.

TRIPADVISOR LLC, 2017. 10 Restoran Terbaik di Tegal - TripAdvisor. [online] Available at: <https://www.tripadvisor.co.id/Restaurants-g790289-Tegal_Central_Java_Java.html> [Accessed 17 Oct. 2017].

TRIPATHY, A., AGRAWAL, A. and RATH, S.K., 2015. Classification of Sentimental Reviews Using Machine Learning Techniques. Procedia Computer Science, [online] 57, pp.821–829. Available at: <http://dx.doi.org/10.1016/j.procs.2015.07.523>.

WANG, S., LI, D., ZHAO, L. and ZHANG, J., 2013. Sample cutting method for imbalanced text sentiment classification based on BRC. Knowledge-Based Systems, [online] 37, pp.451–461. Available at: <http://dx.doi.org/10.1016/j.knosys.2012.09.003>.

WIJOYO, S.H., HERLAMBANG, A.D., ROZI, F. and ISANTA, S.A., 2017. Optimasi Suffix Tree Clustering dengan Wordnet dan Named Entity Recognition untuk Pengelompokan Dokumen. Jurnal Teknologi Informasi dan Ilmu Komputer, [online] 4(4), p.263. Available at: <http://jtiik.ub.ac.id/index.php/jtiik/article/view/400>.

ZHANG, Z., YE, Q., ZHANG, Z. and LI, Y., 2011. Sentiment classification of Internet restaurant reviews written in Cantonese. Expert Systems with Applications, [online] 38(6), pp.7674–7682. Available at: <http://dx.doi.org/10.1016/j.eswa.2010.12.147>.

Diterbitkan

30-10-2018

Terbitan

Bagian

Ilmu Komputer

Cara Mengutip

Support Vector Machine Berbasis Feature Selection Untuk Sentiment Analysis Kepuasan Pelanggan Terhadap Pelayanan Warung dan Restoran Kuliner Kota Tegal. (2018). Jurnal Teknologi Informasi Dan Ilmu Komputer, 5(5), 537-548. https://doi.org/10.25126/jtiik.201855867