KLASIFIKASI TWEET BENCANA MENGGUNAKAN HYBRID CONVOLUTIONAL NEURAL NETWORK DAN GATED RECURRENT UNIT

• RICKO ANUGRAH MULYA PRATAMA
• 14207093

---

ABSTRAK

 

ABSTRAK
Sistem pemantauan bencana dengan menggunakan data Twitter dapat
memberikan informasi terkait daerah rawan bencana, informasi tanggap darurat,
informasi bantuan, dan informasi korban bencana. Terjadinya bencana alam sulit
dicegah, oleh karena itu banyak organisasi bantuan bencana dan berita tertarik
untuk memantau informasi bencana dari Twitter secara terprogram. Informasi ini
sangat penting bagi tim tanggap bencana, karena dapat memberikan informasi
secara real time untuk menentukan tindakan yang tepat dalam mitigasi bencana.
Ada beberapa penelitian yang bertujuan untuk mengimplementasikan pembelajaran
mesin dan teknologi pembelajaran mendalam untuk secara otomatis mendeteksi
informasi bencana dari data Twitter. Salah satu algoritma yang banyak diterapkan
untuk kasus klasifikasi teks adalah Support Vector Machine (SVM), namun SVM
memiliki keterbatasan untuk skala dataset yang besar seperti dataset twitter. Selain
itu, pendekatan deep learning yang umum digunakan untuk klasifikasi teks adalah
Long Short Term Memory (LSTM), namun proses kerja LSTM menggunakan tahap
yang cukup panjang sehingga memerlukan waktu komputasi yang lebih lama.
Gagasan utama dalam mengusulkan model hybrid ini adalah untuk menggabungkan
keunggulan arsitektur Convolutional Neural Network (CNN) yang sangat handal
untuk menangani data berdimensi tinggi dan Gated Recurrent Unit (GRU) yang
efektif dalam mengolah data sequence dan memiliki waktu komputasi yang lebih
cepat dibandingkan dengan LSTM. Dengan masing-masing keunggulannya,
kombinasi tersebut diharapkan dapat menghasilkan model klasifier yang optimal
untuk klasifikasi tweet bencana. Penelitian ini menggunakan GloVe dan FastText
sebagai representasi teks data tweet. Keduanya diuji terhadap model yang diusulkan
menggunakan dataset NLP Disaster Tweets dari forum Kaggle. Hasil kinerja model
yang diusulkan mengungguli setidaknya 12 jenis algoritma machine learning
klasik. Selain itu, model hybrid CNN-GRU juga menghasilkan kinerja yang lebih
baik jika dibandingkan dengan model deep learning yang umum seperti CNN,
LSTM, dan GRU. Model hybrid CNN-GRU dengan teknik word embedding
FastText mampu menghasilkan skor akurasi sebesar 83,32%, skor F1 sebesar
81,45%, dan skor AUC mencapai 83,45%.
Kata kunci:
Bencana, Twitter, Klasifikasi, Machine Learning, Hybrid CNN-GRU
 

KATA KUNCI

Klasifikasi,Hybrid Convolutional Neural Network,Gated Recurrent Unit

---

DAFTAR PUSTAKA

 

DAFTAR PUSTAKA
[1] Ketmaneechairat, H. and Maliyaem, M., "Natural language processing for
disaster management using conditional random fields," Journal of Advances
in Information Technology., vol. 11, no. 2, 2020.
[2] Fan, C., Zhang, C., Yahja, A. and Mostafavi, A., "Disaster City Digital Twin:
A vision for integrating artificial and human intelligence for disaster
management," International Journal of Information Management., vol. 56, p.
102049, 2021.
[3] Ahuja, R., Chug, A., Kohli, S., Gupta, S. and Ahuja, P., "The impact of
features extraction on the sentiment analysis," Procedia Computer Science.,
vol. 152, pp. 341-348, 2019.
[4] R. Collobert, J. Weston, L. Bottou, M. Karlen, K. Kavukcuoglu, and P.
Kuksa, “Natural language processing (almost) from scratch,” Journal of
machine learning research, vol. 12, no. ARTICLE, pp. 2493–2537, 2011.
[5] Khan, N.S., Abid, A. and Abid, K., "A novel natural language processing
(NLP)-based machine translation model for English to Pakistan sign
language translation," Cognitive Computation., vol. 12, no. 4, pp. 748-765,
2020.
[6] Ameur, M.S.H., Belkebir, R. and Guessoum, A., "Robust arabic text
categorization by combining convolutional and recurrent neural networks,"
ACM Transactions on Asian and Low-Resource Language Information
Processing (TALLIP)., vol. 19, no. 5, pp. 1-16, 2020.
[7] Hahn, U. and Oleynik, M., "Medical information extraction in the age of
deep learning," Yearbook of medical informatics., vol. 29, no. 1, pp. 208-
220, 2020.
[8] Boorugu, R. and Ramesh, G., "A survey on NLP based text summarization
for summarizing product reviews," 2nd International Conference on
Inventive Research in Computing Applications., pp. 352-356, 2020.
[9] Van Zanten, G.V., Bouma, G., Sima'an, K., van Noord, G. and Bonnema, R.,
"Evaluation of the NLP components of the OVIS2 spoken dialogue system,"
Computational Linguistics in the Netherlands 2000., pp. 213-229, 2021.
[10] Kowsari, K., Jafari Meimandi, K., Heidarysafa, M., Mendu, S., Barnes, L.,
and Brown, D, “Text classification algorithms: A survey,” Information, vol.
10, no. 4, p. 150, 2019.
[11] Zhang, Y., Liu, Q., and Song, L, “Sentence-state lstm for text
representation,” arXiv preprint arXiv:1805.02474, 2018.
[12] Purnama, I.K.E. and Zaini, A., “Pengembangan agent antarmuka cerdas
berbasis bahasa alami untuk bahasa Indonesia yang diterapkan pada game
edukasi kecakapan hidup (Life Skill),” ITS Digital Repository, 2013.
[13] Cohen, K.B., “Natural language processing for online applications: Text
retrieval, extraction and categorization,” Language Journal, vol. 80, no. 1,
p.178, 2004.
43
[14] Girnanfa FA, Susilo A, “Studi Dramaturgi Pengelolaan Kesan Melalui
Twitter Sebagai Sarana Eksistensi Diri Mahasiswa di Jakarta,” Journal of
New Media and Communication, vol. 1, no.1, pp. 58–73, 2022.
[15] Carter, N, “Disaster Management: A Disaster Manager’s Handbook”, 2008.
[16] UNISDR, “2009 UNISDR Terminology on Disaster Risk Reduction”,
Retrieved from: http://www.unisdr.org/files/7817_UNISDR
TerminologyEnglish.pdf, [Accessed: 12-Des-2022].
[17] Danukusumo, K. P., Pranowo, & Maslim, M., “Indonesia ancient temple
classification using Convolutional Neural Network,” In 2017 International
Conference on Control, Electronics, Renewable Energy and
Communications (ICCREC), pp. 50–54, 2017.
[18] Wolfewicz, A., “Deep Learning vs. Machine Learning - What’s The
Difference?,” 2022 Retrieved from: https://levity.ai/blog/differencemachine-learning-deep-learning, [Accessed: 12-Des-2022].
[19] L. M. Azizah, S. F. Umayah, and F. Fajar, “Deteksi kecacatan permukaan
buah manggis menggunakan metode deep learning dengan konvolusi
multilayer,” Semesta Teknika, vol. 21, no. 2, pp. 230–236, 2018.
[20] K. Fukushima, "Neocognitron: A Self-Organizing Neural Network Model
for a Mechanism of Pattern Recognition Unaffected by Shift in Position,"
Biological Cybernetics, 1980.
[21] Y. LeCun, "Handwritten Digit Recognition with a BackPropagation
Network," 1990.
[22] MatWorks, “Speech Command Recognition Using Deep Learning,”
Retrieved from: https://www.mathworks.com/discovery/convolutionalneural-network-matlab.html, [Accessed: 12-Des-2022].
[23] P. Benny, “Student Notes: Convolutional Neural Networks Introduction,”
Retrieved from: https://indoml.com/2018/03/07/student-notesconvolutional-neuralnetworks-cnn-introduction/, [Accessed: 12-Des-2022].
[24] J. Brownlee, “A Gentle Introduction to the Rectified Linear Unit,” Retrieved
from: https://machinelearningmastery.com/rectified-linear-activationfunction-for-deep-learning-neural-networks/, [Accessed: 12-Des-2022].
[25] Arc, “An Introduction to Convolutional Neural Networks,” Retrieved from:
https://towardsdatascience.com/convolutional-neural-network-
17fb77e76c05/, [Accessed: 12-Des-2022].
[26] Y. R. Aditya, “Fully-Connected Layer CNN dan Implementasinya,”
Retrieved from: https://socs.binus.ac.id/2017/02/13/rnn-dan-gru/,
[Accessed: 12-Des-2022].
[27] K. Cho, D. Bahdanau, F. Bougares, H. Schwenk, dan Y. Bengio, “Learning
Phrase Representations using RNN Encoder-Decoder for Statistical Machine
Translation,” arXiv:1406.1078 [cs.CL], 2014.
[28] P. Aryo, S. Derwin, “Recurrent Neural Network dan Gated Recurrent Unit,”
Retrieved from: https://socs.binus.ac.id/2017/02/13/rnn-dan-gru/,
[Accessed: 12-Des-2022].
[29] D. Saul, “GRU Recurrent Neural Networks-A Smart Way to Predict
Sequences in Python,” Retrieved from: https://towardsdatascience.com/gru
44
recurrent-neural-networks-a-smartway-to-predict-sequences-in-python-
80864e4fe9f6/, [Accessed: 12-Des-2022].
[30] Nurdin, A., Aji, B. A. S., Bustamin, A., and Abidin, Z., “Perbandingan
Kinerja Word Embedding Word2Vec, Glove, Dan Fasttext Pada Klasifikasi
Teks,” Jurnal Tekno Kompak, vol. 14, no. 2, pp. 72–79, 2020.
[31] J. Pennington, R. Socher, and C. D. Manning, “Glove: Global vectors for
word representation,” in Proceedings of the 2014 conference on empirical
methods in natural language processing (EMNLP), pp. 1532–1543, 2014.
[32] Boukkouri, H.E., “Arithmetic Properties of Word Embeddings”, 2020
Retrieved from: https://blog.dataiku.com/arithmetic-properties-of-wordembeddings, [Accessed: 12-Des-2022].
[33] Heaton, J., “Introduction to the Math of Neural Networks (Beta-1),” Heaton
Research Inc., 2011.
[34] P. Baheti, “Activation Functions in Neural Networks [12 Types & Use
Cases],” Retrieved from: https://www.v7labs.com/blog/neural-networksactivation-functions, [Accessed: 12-Des-2022].
[35] Jain, A., Shakya, A., Khatter, H. and Gupta, A.K., "A smart system for fake
news detection using machine learning," 2019 International Conference on
Issues and Challenges in Intelligent Computing Techniques (ICICT)., vol. 1,
pp. 1-4, 2019.
[36] Villavicencio, C., Macrohon, J.J., Inbaraj, X.A., Jeng, J.H. and Hsieh, J.G.,
"Twitter sentiment analysis towards covid-19 vaccines in the Philippines
using naïve bayes," Information., vol. 12, no. 5, p. 204, 2021.
[37] Ginting, P.S.B., Irawan, B. and Setianingsih, C., "Hate speech detection on
Twitter using multinomial logistic regression classification method," 2019
IEEE International Conference on Internet of Things and Intelligence System
(IoTaIS)., pp. 105-111, 2019.
[38] Dharma, E.M., Gaol, F.L., Leslie, H., Warnars, H.S. and Soewito, B., "The
accuracy comparison among word2vec, glove, and fasttext towards
convolution neural network text classification," Journal of Theoretical and
Applied Information Technology., vol. 100, no. 2, p. 31, 2022.
[39] Marpaung, A., Rismala, R. and Nurrahmi, H., "Hate Speech Detection in
Indonesian Twitter Texts using Bidirectional Gated Recurrent Unit," 2021
13th International Conference on Knowledge and Smart Technology (KST).,
pp. 186-190, 2021.
[40] Ajao, O., Bhowmik, D. and Zargari, S., "Fake news identification on twitter
with hybrid cnn and rnn models," In Proceedings of the 9th international
conference on social media and society., pp. 226-230, 2018.
[41] Jain, P.K., Saravanan, V. and Pamula, R., " A hybrid CNN-LSTM: A deep
learning approach for consumer sentiment analysis using qualitative user
generated contents," Transactions on Asian and Low-Resource Language
Information Processing., vol. 20, no. 5, pp. 1-15, 2021.
[42] Nistor, S.C., Moca, M., Moldovan, D., Oprean, D.B. and Nistor, R.L.,
“Building a Twitter sentiment analysis system with recurrent neural
networks,”, Sensors., vol. 21, no. 7, p. 2266, 2021.
45
[43] Venkataramaiah, M.K.A. and Achar, N.A.N., “Twitter sentiment analysis
using aspect-based bidirectional gated recurrent unit with self-attention
mechanism,” International Journal of Intelligent Engineering and Systems.,
vol. 13, no. 5, pp. 97-110, 2020.
[44] Edo-Osagie, O., Lake, I., Edeghere, O. and Iglesia, B.D.L., “Attention-based
recurrent neural networks (RNNs) for short text classification: An
application in public health monitoring,” international work-conference on
artificial neural networks, pp. 895–911, 2019.
[45] Kaggle, “Natural Language Processing with Disaster Tweets,” Retrieved Oct
15, 2022 from https://www.kaggle.com/c/nlp-getting-started.
[46] Pradany, L.N. and Fatichah, C., “Analisa Sentimen Kebijakan Pemerintah
Pada Konten Twitter Berbahasa Indonesia Menggunakan Svm Dan KMedoid Clustering,” SCAN-Jurnal Teknologi Informasi dan Komunikasi.,
vol. 11, no. 1, pp. 59-66, 2016.
[47] Nurjaman, Janjan and Ilyas, Ridwan and Kasyidi, Fatan, “Pengukuran
Kesamaan Semantik Pasangan Kalimat Sitasi Menggunakan Convolutional
Neural Network,” Prosiding Industrial Research Workshop and National
Seminar, vol. 11, no. 1, pp. 510–516, 2020.
[48] L. Francis dan M. Flynn., “Text Mining Handbook,” Casualty Actuarial
Society E-Forum, 2010.
[49] Ihsan, M., Negara, B.S. and Agustian, S, “LSTM for Sentiment COVID-19
Vaccine Classification on Twitter,” Digital Zone: Jurnal Teknologi
Informasi Dan Komunikasi, vol. 13, no. 1, pp. 79–89, 2022.
[50] Nasir, M.N., “Perbandingan pengaruh nilai centroid awal pada algoritma KMeans dan K-Means++ terhadap hasil cluster menggunakan metode
confusion matrix,” Soliter, vol. 1, pp. 118–127, 2018.
[51] F. Gorunescu, “Data Mining: Concepts, models and techniques,” Springer
Science & Business Media, vol. 12, 2011.