Aspect-Based Sentiment Analysis pada Ulasan ECommerce dengan Metode Support Vector Machine untuk Mendapatkan Informasi Sentimen dari Beberapa Aspek

Authors

  • Hansen Gunawan Sulistio Program Studi Teknik Informatika, Universitas Kristen Petra Surabaya
  • Andreas Handojo Program Studi Teknik Informatika, Universitas Kristen Petra Surabaya

Keywords:

OHS management system, SMK3, SMK3 design, PP nomor 50 tahun 2012

Abstract

In this era of globalization, all people's activities are starting to use technology to facilitate their daily activities. One of the most impactful forms of digitization activities is online buying and selling activities such as the use of the Tokopedia and Shopee platforms. The existence of a review feature on online buying and selling places (e-commerce) is one of the factors supporting the increase in people's online transactions. The number of people who have started to implement online buying and selling activities in their daily lives has resulted in an increase in the number of reviews on e-commerce.The large number of reviews makes it difficult for potential buyers to review a product to be purchased. The factors that determine the shopping experience of each individual are different so that the reviews and ratings given by each individual on a product or store vary. This affects the average rating of a product or store so that the average rating on a product does not necessarily represent the quality of the product. To overcome this problem, the author makes a system where prospective buyers and sellers will be facilitated to assess an aspect of a product. The system created by the author shows several aspects that are crucial for buyers and sellers in reading a review, such as general aspects, accuracy, quality, service, delivery, packaging, and price using the Support Vector Machine method. In these aspects, the system created will show sentiments on reviews that have been written by buyers such as positive, negative, or neutral sentiments. In addition to showing the sentiments of aspects of a product, this system also shows which aspects affect the product rating the most, the aspects that are most frequently discussed, what aspects are most rated positively and negatively.The results of the thesis show that the aspect that is often discussed is the quality aspect. General aspects, accuracy, quality, service, delivery, and packaging affect the rating value on a product rating while the price aspect does not affect a product rating. Compared to the Shopee platform, there are more positive reviews written on Tokopedia than reviews on Shopee.

References

[1] Cortes, C., & Vapnik, V. (1995). Support-vector networks.

Machine learning, 20(3), 273-297.doi:

https://doi.org/10.1007/BF00994018

[2] Devi, D. N., Kumar, C. K., & Prasad, S. (2016, February). A

feature based approach for sentiment analysis by using

support vector machine. In 2016 IEEE 6th International

Conference on Advanced Computing (IACC), 3-8. doi:

10.1109/IACC.2016.11.

[3] Finkel, J. R., & Manning, C. D. (2009, August). Nested

named entity recognition. In Proceedings of the 2009

conference on empirical methods in natural language

processing, 141-150.

[4] Joachims, T. (1998). Text categorization with support vector

machines: Learning with many relevant features. European

Conference on Machine Learning, 1398, 137-142.

https://doi.org/10.1007/BFb0026683

[5] Lu, Y., Zhai, C., & Sundaresan, N. (2009). Rated aspect

summarization of short comments. WWW '09: Proceedings

of the 18th international conference on World wide web,

131–140. https://doi.org/10.1145/1526709.1526728

[6] Raja, K., & Pushpa, S. (2017). Feature level review table

generation for E-Commerce websites to produce qualitative

rating of the products. Future Computing and Informatics

Journal, 2, 118-124.

https://doi.org/10.1016/j.fcij.2017.09.002

[7] Riany, J., Fajar, M., & Lukman, M. P. (2016). Penerapan

deep sentiment analysis pada angket penilaian terbuka

menggunakan K-Nearest Neighbor. SISFO Vol 6 No 1, 6.

doi: https://doi.org/10.24089/j.sisfo.2016.09.011

[8] Santra, A., & Christy, J. (2012). Genetic Algorithm and

Confusion Matrix for Document Clustering. IJCSI

International Journal of Computer Science Issues, 9(1), 322-

328.

[9] Septiadi, A., & Ramadhani, W. K. (2020). Penerapan Metode

Anova untuk Analisis Rata-rata Produksi Donat, Burger, dan

Croissant pada Toko Roti Animo Bakery. Bulletin of Applied

Industrial Engineering Theory, 1(2720-9628), 60-64.

[10] Setiawan, G., Palit, H., & Setyati, E. (2019). Aspect Based

Sentiment Analysis pada Layanan Umpan Balik Universitas

dengan Menggunakan Metode Naïve Bayes dan Latent

Semantic Analysis. JURNAL INFRA, 7(1), 170-174.

[11] Somantri, O., & Apriliani, D. (2018). Support Vector

Machine Berbasis Feature Selection Untuk Sentiment

Analysis Kepuasan Pelanggan Terhadap Pelayanan Warung

Dan Restoran Kuliner Kota Tegal. Jurnal Teknologi

Informasi dan Ilmu Komputer, 5(2355-7699), 537-547.

10.25126/jtiik20185867

[12] Tala, F. (2003). A study of stemming effects on information

retrieval in Bahasa Indonesia.

[13] Wahyudi, D., Susyanto, T., & Nugroho, D. (2017).

Implementasi Dan Analisis Algoritma Stemming Nazief &

Adriani Dan Porter Pada Dokumen Berbahasa Indonesia.

Jurnal Ilmiah SINUS, 15(2).

http://dx.doi.org/10.30646/sinus.v15i2.305

[14] Zhang, J., Jin, R., Yang, Y., & Hauptmann, A. (2003).

Modified logistic regression: An approximation to SVM and

its applications in large-scale text categorization. 888-895

Downloads

Published

2022-08-29

Issue

Section

Articles