Implementasi Machine Learning untuk Analisis Sentimen Opini Publik Mengenai Pembukaan Kebun Kelapa Sawit di Papua: Machine Learning-Based Sentiment Analysis of Public Opinion on Palm Oil Plantation Expansion in Papua

Ester Ayuk Pusvita; Deni Stefanus Paboy Ranggup; Usman Arfan

doi:10.57152/malcom.v5i4.2279

Authors

Ester Ayuk Pusvita STMIK Pesat Nabire
Deni Stefanus Paboy Ranggup STMIK Pesat Nabire
Usman Arfan STMIK Pesat Nabire

DOI:

https://doi.org/10.57152/malcom.v5i4.2279

Keywords:

Analisis Sentimen, Emosi Publik, Machine Learning, Multilingual Sentiment, Papua

Abstract

Ekspansi perkebunan kelapa sawit di Papua menimbulkan beragam respons publik di media sosial, khususnya pada platform X (Twitter). Penelitian ini bertujuan untuk menganalisis sentimen dan emosi publik terhadap isu tersebut menggunakan pendekatan machine learning melalui aplikasi Orange Data Mining. Sebanyak 1.355 tweet dikumpulkan menggunakan Twitter API v2 dan Google Colab yang terintegrasi dengan pustaka snscrape guna mengatasi batasan pengambilan data. Analisis dilakukan dengan model Multilingual Sentiment untuk klasifikasi polaritas (positif, negatif, netral) dan model Ekman Emotion untuk identifikasi enam emosi dasar. Hasil menunjukkan bahwa sentimen netral mendominasi (39,78%), diikuti negatif (32,16%) dan positif (28,05%). Namun, sentimen netral tidak selalu bersifat informatif, melainkan dapat muncul akibat ambiguitas linguistik atau keterbatasan model dalam memahami konteks lokal bahasa Indonesia. Emosi Joy (sukacita) merupakan emosi paling dominan, tetapi juga muncul dalam kategori sentimen negatif, yang mengindikasikan adanya ekspresi sarkasme dan ironi terhadap isu sawit. Hal ini mencerminkan keterbatasan model otomatis dalam mendeteksi makna tersirat dan gaya bahasa satir. Penelitian ini menyimpulkan bahwa meskipun model Multilingual Sentiment efektif untuk mendeteksi pola umum opini publik, pendekatan ini memerlukan penyesuaian kontekstual agar lebih sensitif terhadap nuansa budaya dan semantik bahasa Indonesia

Downloads

Download data is not yet available.

References

K. T. Sibhatu, “Oil palm boom: its socioeconomic use and abuse,” Front. Sustain. food Syst., vol. 7, 2023, doi: 10.3389/fsufs.2023.1083022.

S. F. Purba et al., “Strategies for improving independent oil palm smallholders’ welfare in Konawe Regency, Southeast Sulawesi,” 2024, doi: 10.1088/1755-1315/1379/1/012013.

R. Adawiyah, Z. Zubir, and H. Efendi, “Perampasan Tanah dan Perlawanan Petani: Dampak Perkebunan Sawit terhadap Kehidupan Masyarakat di Pasaman Barat Tahun 1980-2022,” Ethnoreflika J. Sos. dan Budaya, vol. 13, no. 1, pp. 1–23, 2024, doi: 10.33772/etnoreflika.v13i1.2429.

Anju, “Exploring the Impact of Social Media on Public Opinion Formation: A Comparative Analysis,” Int. J. Res. Appl. Sci. Eng. Technol., 2024, doi: 10.22214/ijraset.2024.59619.

V. J. de Castro Paes, D. V. G. Araújo, K. Brito, and E. Andrade, “Análise de Sentimento em Tweets Relacionados ao Desmatamento da Floresta Amazônica,” 2022, doi: 10.5753/brasnam.2022.222648.

T. Kumaragurubaran, S. Pandi, G. Naresh, and T. S. Ragavender, “Navigating Public Opinion: Enhancing Sentiment Analysis on Social Media with CNN and SVM,” 2024, doi: 10.1109/incet61516.2024.10592999.

L. Yu, “Public Opinion Monitoring of Sports Stars Based on Text Sentiment Analysis,” Int. J. Comput. Sci. Inf. Technol., 2024, doi: 10.62051/ijcsit.v4n2.02.

U. Arfan and N. Paraga, “Perbandingan Algoritma K-Means, Na{"i}ve Bayes dan Decision Tree Dalam Memprediksi Penjualan Bahan Bakar Minyak: The Comparison of K-Means, Na{"i}ve Bayes and Decision Tree Algorithm in Predicting Fuel Oil Sales,” MALCOM Indones. J. Mach. Learn. Comput. Sci., vol. 4, no. 4, pp. 1379–1389, 2024.

A. A. Mammadov and G. Bakal, “From Traditional to Deep: Evaluating Sentiment Analysis Models on a Large-Scale Tweet Dataset,” pp. 1–6, 2024, doi: 10.1109/ubmk63289.2024.10773489.

K. Wangi and A. C. Inamdar, “Machine Learning Techniques For Twitter Data Using Sentimental Analysis,” pp. 1–5, 2024, doi: 10.1109/innova63080.2024.10846953.

K. K. Kumar et al., “Text Classification on Twitter Data Using Machine Learning Algorithm,” Int. J. Res. Appl. Sci. Eng. Technol., 2023, doi: 10.22214/ijraset.2023.57419.

R. H. Dwijayani, M. A. S. Ali, and S. Sugito, “Hutan industri dan deforestasi: bagaimana hutan industri mengancam keberlangsungan hutan hujan di papua, indonesia,” J. Agrifor, vol. 22, no. 2, p. 233, 2023, doi: 10.31293/agrifor.v22i2.6719.

A. M. Gomez, A. Parra, T. M. Pavelsky, E. K. Wise, J. C. Villegas, and A. Meijide, “Ecohydrological impacts of oil palm expansion: a systematic review,” Environ. Res. Lett., vol. 18, no. 3, p. 33005, 2023, doi: 10.1088/1748-9326/acbc38.

H. Lawelai and A. Sadat, “Trend Analysis of Positive Sentiment for Special Autonomy for Papua on Twitter,” J. Bina Praja J. Home Aff. Gov., vol. 14, no. 2, pp. 213–224, 2022, doi: 10.21787/jbp.14.2022.213-224.

S. de Oliveira Gonçalves, L. Silveira, and J. I. da Silva Filho, “Aprendizado de máquina sem matemática e programação: um relato de experiência de uma abordagem utilizando o software Orange Data Mining para alunos de administração,” GeSec, vol. 15, no. 4, p. e3700, 2024, doi: 10.7769/gesec.v15i4.3700.

J. Demšar and B. Zupan, “Hands-on training about data clustering with orange data mining toolbox,” PLOS Comput. Biol., vol. 20, no. 12, p. e1012574, 2024, doi: 10.1371/journal.pcbi.1012574.

A. K. Pandey, D. Raghav, G. Gupta, and V. Srivastava, “A Deep Learning Approach for Multiclass Orange Disease Classification,” pp. 184–189, 2024, doi: 10.1109/icdt61202.2024.10489557.

A. M. Kafiar and S. Supatman, “Analisis sentimen netizen terhadap isu pembabatan hutan adat papua melalui tagar #alleyesonpapua menggunakan algoritma support vector machine,” JATI (Jurnal Mhs. Tek. Inform., vol. 8, no. 4, pp. 8129–8135, 2024, doi: 10.36040/jati.v8i4.10672.

Y. N. Bisoumi, J. Munandar, S. Amrullah, M. T. Pandiriyan, K. R. Akmalia, and F. Fauzi, “Papua dalam Perspektif Komentar Youtube: Studi Pemodelan Topik dan Analisis Sentimen dengan Pendekatan Text Mining,” Pros. Semin. Nas. Sains Data, vol. 4, no. 1, pp. 270–281, 2024, doi: 10.33005/senada.v4i1.190.

D. H. Azahari, S. Sukarman, and B. W. van Assen, “Palms of paradox – cultivating palms to support reforestation and avoid deforestation,” vol. 1407, p. 12020, 2024, doi: 10.1088/1755-1315/1407/1/012020.

X.-F. Liu, “A Machine Learning Framework for Document Classification by Topic Recognition Using Latent Dirichlet Allocation and Domain Knowledge,” 2022, pp. 509–520. doi: 10.1007/978-981-19-2821-5_42.

V. Segarra-Faggioni, R. Sylvie, and J. F. Frank, “Topic Modelling for Automatically Identification of Relevant Concepts Discussed in Academic Documents,” Springer International Publishing, 2023, pp. 85–95. doi: 10.1007/978-3-031-33261-6_8.

P. Obiorah, F. E. Onuodu, and B. Eke, “Topic Modeling Using Latent Dirichlet Allocation & Multinomial Logistic Regression,” Adv. Multidiscip. Sci. Res. J., vol. 10, no. 4, pp. 99–112, 2022, doi: 10.22624/aims/digital/v10n4p11a.