Claim Missing Document
Check
Articles

Found 6 Documents
Search

Perbandingan Algoritma Support Vector Machine dan Random Forest untuk Analisis Sentimen Terhadap Kebijakan Pemerintah Indonesia Terkait Kenaikan Harga BBM Tahun 2022 Samantri, Muhamad; Afiyati
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 8 No 1 (2024): JANUARY-MARCH 2024
Publisher : Lembaga Otonom Lembaga Informasi dan Riset Indonesia (KITA INFO dan RISET)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v8i1.1202

Abstract

The commodity of fuel oil (BBM) is the main commodity and the driving force of business. The increase in world oil prices is a threat to countries around the world, one of which is Indonesia. With the turbulent conditions in several countries, the Indonesian government decided to cut fuel subsidies which had an impact on price increases. The policy invited all Indonesian people and criticized it on various social media. The purpose of this research is to find out which algorithm has a better accuracy rate and to provide input to the government about public opinion regarding the increase in fuel prices in Indonesia. From the test results both work well, this is evidenced by the accuracy value obtained, where the support vector machine algorithm produces an accuracy value of 77%, while the Random Forest algorithm produces an accuracy value of 76%. So it can be concluded that the support vector machine algorithm has a fairly good accuracy rate compared to the Random Forest algorithm.
Analisis Sentimen Proyek Strategis Nasional Food Estate Menggunakan Algoritma Naïve Bayes, Logistic Regression dan Support Vector Machine Mustopo, Yuning Rum Zattayu; Afiyati
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 9 No 2 (2025): APRIL-JUNE 2025
Publisher : Lembaga Otonom Lembaga Informasi dan Riset Indonesia (KITA INFO dan RISET)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v9i2.3312

Abstract

The National Strategic Project Food Estate is an initiative by the Indonesian government aimed at enhancing food security through the development of large-scale agricultural areas. In the vice-presidential debate ahead of the 2024 election, Food Estate re-emerged as a hot topic, sparking controversy. Therefore, this study aims to analyze public perspectives on the National Strategic Project Food Estate by comparing the performance of machine learning algorithms, including Naïve Bayes, Logistic Regression and Support Vector Machine. This research also experiments with feature extraction techniques TF-IDF and Word2Vec. The results indicate that TF-IDF feature extraction performs better in capturing relevant features to enhance classification performance compared to the Word2Vec method. The best-performing algorithm is Logistic Regression + TF-IDF, achieving an accuracy of 74%, followed by SVM + TF-IDF and Naïve Bayes + TF-IDF with accuracies of 73% and 72%, respectively.
Prediksi Nilai Unburned Carbon Batubara yang Dihasilkan PLTU Menggunakan Algoritma Linear Regression, Random Forest, dan LightGBM Regression Syarif, Muhyiddin; Afiyati
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 9 No 2 (2025): APRIL-JUNE 2025
Publisher : Lembaga Otonom Lembaga Informasi dan Riset Indonesia (KITA INFO dan RISET)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v9i2.3313

Abstract

This study focuses on predicting unburned carbon levels in coal-fired power plants to enhance operational efficiency. Accurate prediction of unburned carbon is crucial as it directly affects fuel combustion efficiency and environmental sustainability. The research compares three machine learning algorithms: Linear Regression, Random Forest, and LightGBM Regression, using performance metrics such as Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE). The results show that LightGBM Regression performs the best, with MAE of 0.31, MAPE of 1.29, and RMSE of 0.38, outperforming the other two models. This model can be further optimized to improve prediction accuracy, contributing to more efficient and environmentally friendly power plant operations. The application of machine learning in this study supports data-driven decision-making in the energy sector.
Analisis Sentimen Proyek Strategis Nasional Food Estate Menggunakan Algoritma Naïve Bayes, Logistic Regression dan Support Vector Machine Mustopo, Yuning Rum Zattayu; Afiyati
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 9 No 2 (2025): APRIL-JUNE 2025
Publisher : Lembaga Otonom Lembaga Informasi dan Riset Indonesia (KITA INFO dan RISET)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v9i2.3312

Abstract

The National Strategic Project Food Estate is an initiative by the Indonesian government aimed at enhancing food security through the development of large-scale agricultural areas. In the vice-presidential debate ahead of the 2024 election, Food Estate re-emerged as a hot topic, sparking controversy. Therefore, this study aims to analyze public perspectives on the National Strategic Project Food Estate by comparing the performance of machine learning algorithms, including Naïve Bayes, Logistic Regression and Support Vector Machine. This research also experiments with feature extraction techniques TF-IDF and Word2Vec. The results indicate that TF-IDF feature extraction performs better in capturing relevant features to enhance classification performance compared to the Word2Vec method. The best-performing algorithm is Logistic Regression + TF-IDF, achieving an accuracy of 74%, followed by SVM + TF-IDF and Naïve Bayes + TF-IDF with accuracies of 73% and 72%, respectively.
Prediksi Nilai Unburned Carbon Batubara yang Dihasilkan PLTU Menggunakan Algoritma Linear Regression, Random Forest, dan LightGBM Regression Syarif, Muhyiddin; Afiyati
Jurnal JTIK (Jurnal Teknologi Informasi dan Komunikasi) Vol 9 No 2 (2025): APRIL-JUNE 2025
Publisher : Lembaga Otonom Lembaga Informasi dan Riset Indonesia (KITA INFO dan RISET)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/jtik.v9i2.3313

Abstract

This study focuses on predicting unburned carbon levels in coal-fired power plants to enhance operational efficiency. Accurate prediction of unburned carbon is crucial as it directly affects fuel combustion efficiency and environmental sustainability. The research compares three machine learning algorithms: Linear Regression, Random Forest, and LightGBM Regression, using performance metrics such as Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE). The results show that LightGBM Regression performs the best, with MAE of 0.31, MAPE of 1.29, and RMSE of 0.38, outperforming the other two models. This model can be further optimized to improve prediction accuracy, contributing to more efficient and environmentally friendly power plant operations. The application of machine learning in this study supports data-driven decision-making in the energy sector.
Sentimen Analisis Pada Ulasan Aplikasi Indodana di Google Play Store Menggunakan Algoritma Logistic Regression, Naive Bayes dan Support Vector Machine Sahrul, Muhammad; Afiyati
Jurnal Nasional Teknologi Komputer Vol 5 No 3 (2025): Juli 2025
Publisher : CV. Hawari

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Penelitian ini bertujuan mengevaluasi sentimen pengguna berdasarkan ulasan aplikasi Indodana di Play Store, menggunakan algoritma Logistic Regression, Naive Bayes, dan Support Vector Machine (SVM). Tujuan utama studi ini adalah mengidentifikasi dan mengklasifikasikan ulasan sentimen positif atau negatif, serta mengevaluasi efektivitas sentimen analisis pada produk Paylater dan Pinjaman Online (Pinjol) untuk melihat potensi ketertarikan pengguna terhadap layanan yang ditawarkan. Selain itu, penelitian ini bertujuan untuk membandingkan keakuratan dan performa masing-masing algoritma, guna menentukan algoritma yang paling efektif dalam menganalisis sentimen. Setelah menerapkan algoritma Logistic Regression menghasilkan akurasi 92,01%, presisi 92,12%, recall 91,87%, dan f1-score 91,96%. Sementara itu, algoritma Naive Bayes, diperoleh hasil dengan hasil akurasi 80,91% , presisi 80,69%, recall 80,36% , dan f1-score 80,54%.Di sisi lain, algoritma SVM mencatatkan nilai akurasi 92,63%, presisi 92,17%, recall 91,98%, dan f1-score 92,06%. Pernyataan ini menunjukkan bahwa dalam penelitian ini, algoritma SVM memiliki performa yang lebih baik dibandingkan Logistic Regression, dan Naive Bayes. Diharapkan hasil penelitian ini dapat memberikan pemahaman yang lebih mendalam dan berguna untuk pengembangan aplikasi Indodana serta aplikasi-aplikasi serupa, dengan membantu perusahaan memahami umpan balik pengguna secara lebih mendetail. Dengan demikian, penelitian ini dapat berkontribusi pada perumusan strategi pemasaran yang lebih tepat sasaran serta peningkatan kualitas produk dan layanan yang lebih baik. Penelitian ini juga diharapkan dapat menjadi referensi untuk penelitian selanjutnya di bidang sentimen analisis pada Paylater dan Pinjaman Online, sehingga dapat memberikan nilai tambah bagi perusahaan-perusahaan yang bergerak di sektor keuangan digital.