Peningkatan Akurasi Klasifikasi Sentimen Ulasan Makanan Amazon dengan Bidirectional LSTM dan Bert Embedding

David Junggu Manggala Pasaribu, Kusrini Kusrini, Sudarmawan Sudarmawan

Abstract


Sudah memasuki revolusi industri 4.0 dengan infrastuktur internet semakin memadai dan biaya lebih murah mengakibatkan banyak masyarakat menggunakan layanan pada internet. Sehingga organisasi bisnis terdisrupsi untuk merambah ke media online. Seperti Amazon perusahaan e-commerce meliputi Costumer to Costumer maupun Business to Business, salah satu produk yang dipasarkan adalah makanan. Untuk menaikkan pemasukannya maka perusahaan harus mengerti kebutuhan pembeli. Sehingga dilakukan analisis sentimen konsumen namun proses ini memerlukan waktu lama sehingga dibuat secara otomatis menggunakan metode kecerdasan buatan. Dalam hasil penelitian tentang analisis sentimen pada dataset Amazon Fine Food Review menggunakan metode deep learning Bidirectional Long Short-Term Memory dengan penghasil vektor kata Bidirectional Encoder Representations from Transformers mampu menghasilkan akurasi yang lebih baik daripada menggunakan smetode machine learning Logistic Regression dengan pembobotan kata Mutual Information dan Bag of Words serta model deep learning Convolutional Neural Network dan Long Short-Term Memory dengan penghasil vektor kata Word2Vec dan GloVe pada konfigurasi ukuran embedding dan jumlah dataset paling besar yaitu 300 dan 85.000 sebesar 93 %.

 


Keywords


amazon fine food review; analisis sentimen; Bidirectional Encoder Representations from Transformers; Bidirectional Long Short-Term Memory; Word2Vec

Full Text:

PDF

References


Anees, Aiman Abdullah et al. 2019. “Performance Analysis of Multiple Classifiers Using Different Term Weighting Schemes for Sentiment Analysis.” 2019 International Conference on Intelligent Computing and Control Systems, ICCS 2019 (Iciccs): 637–41.

Buduma, Nikhil. 2017. Fundametals of Deep Learning : Designing Next-Generation Machine Intelligence Algorithms. Sebastopol: O’Reilly Media, Inc.

Chemchem, Amine, François Alin, and Michael Krajecki. 2019. “Improving the Cognitive Agent Intelligence by Deep Knowledge Classification.” International Journal of Computational Intelligence and Applications 18(1): 1–25.

Devlin, Jacob, Ming Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. “BERT: Pre-Training of Deep Bidirectional Transformers for Language Understanding.” NAACL HLT 2019 - 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Proceedings of the Conference 1(Mlm): 4171–86.

Graves, Alex, Navdeep Jaitly, and Abdel-rahman Mohamed. 2013. “HYBRID SPEECH RECOGNITION WITH DEEP BIDIRECTIONAL LSTM Alex Graves , Navdeep Jaitly and Abdel-Rahman Mohamed University of Toronto Department of Computer Science 6 King ’ s College Rd . Toronto , M5S 3G4 , Canada.” : 273–78.

Kingma, Diederik P., and Jimmy Lei Ba. 2015. “Adam: A Method for Stochastic Optimization.” 3rd International Conference on Learning Representations, ICLR 2015 - Conference Track Proceedings: 1–15.

Maas, Andrew L. et al. 2011. “Learning Word Vectors for Sentiment Analysis.” The 49th Annual Meeting of the Association for Computational Linguistics.

Meeker, Mary. 2019. “BOND Internet Trends 2019.” 2019-06-11: 333. https://www.bondcap.com/report/itr19/#view/1.

Prabhavathi, C. 2019. “Machine Learning Model for Classifying L _ Text Using Nlp ( Amazon Product Reviews ).” 6(04): 161–78.

Ruder, Sebastian. 2016. “An Overview of Gradient Descent Optimization Algorithms.” : 1–14. http://arxiv.org/abs/1609.04747.

Suyanto, K. N. Ramadhani, and S. Mandala. 2019. Deep Lerning: Modernisasi Machine Learning Untuk Big Data. Bandung: Informatika.

Zhou, Zhenxiang, and Lan Xu. 2016. “Amazon Food Review Classification Using Deep Learning and Recommender System.” Stanford University: 1–7. https://cs224d.stanford.edu/reports/ZhouXu.pdf.


DOI http://dx.doi.org/10.35585/inspir.v10i1.2568
Abstract 74 kali dilihat
PDF 58 kali diunduh

Refbacks

  • There are currently no refbacks.



Lisensi Creative Commons
Ciptaan disebarluaskan di bawah Lisensi Creative Commons Atribusi-NonKomersial-BerbagiSerupa 4.0 Internasional.