Abstract:Penggunaan dompet digital yang terus meningkat menghasilkan banyak ulasan pengguna yang dapat dimanfaatkan untuk mengevaluasi kualitas layanan. Penelitian ini bertujuan meningkatkan akurasi klasifikasi sentimen pengguna…
dompet digital menggunakan metode Stacking Ensemble Machine Learning yang mengombinasikan Support Vector Machine (SVM), Multinomial Naïve Bayes (MNB), dan AdaBoost dengan Logistic Regression sebagai meta-learner. Data ulasan diproses melalui tahapan text preprocessing meliputi case folding, cleaning, tokenizing, stopword removal, stemming, dan pembobotan fitur menggunakan TF-IDF. Penyeimbangan data dilakukan dengan SMOTE, sedangkan evaluasi model menggunakan 5-Fold Cross-Validation. Hasil penelitian menunjukkan bahwa model Stacking Ensemble memperoleh akurasi rata-rata 80,55%, lebih tinggi dibandingkan algoritma dasar. Evaluasi menggunakan Confusion Matrix, Classification Report, dan ROC Curve juga menunjukkan peningkatan nilai precision, recall, F1-score, dan kemampuan diskriminasi model. Hasil ini menunjukkan bahwa pendekatan Stacking Ensemble Machine Learning efektif untuk meningkatkan akurasi klasifikasi sentimen pengguna dompet digital serta mendukung evaluasi kualitas layanan berbasis opini pengguna.
The rapid growth of digital wallet usage has generated a large volume of user reviews that can be utilized to evaluate service quality. This study aims to improve the accuracy of digital wallet user sentiment classification using a Stacking Ensemble Machine Learning approach that combines Support Vector Machine (SVM), Multinomial Naïve Bayes (MNB), and AdaBoost with Logistic Regression as the meta-learner. User reviews were processed through text preprocessing stages, including case folding, text cleaning, tokenization, stopword removal, stemming, and TF-IDF feature weighting. Synthetic Minority Over-sampling Technique (SMOTE) was employed to address class imbalance, while model performance was evaluated using 5-Fold Cross-Validation. The experimental results show that the proposed Stacking Ensemble model achieved an average accuracy of 80.55%, outperforming the individual base learners. Furthermore, evaluations based on the Confusion Matrix, Classification Report, and Receiver Operating Characteristic (ROC) Curve demonstrated improvements in precision, recall, F1-score, and the model's discriminative capability. These findings indicate that the proposed Stacking Ensemble Machine Learning approach is effective in improving the accuracy of digital wallet user sentiment classification and can serve as a reliable tool for supporting service quality evaluation based on user opinions.
Abstract:PayLater services are one of the rapidly growing digital financial innovations widely utilised in fintech apps in Indonesia, including Kredivo and Akulaku. User reviews on the Google Play Store reflect a range of experiences,…
nces, from satisfaction with the ease of use of the service to complaints regarding bills, interest rates, late payment fees, credit limits, and app performance. This study aims to classify the sentiment of user reviews regarding PayLater services on the Kredivo and Akulaku apps using the Multinomial Naïve Bayes algorithm. Data was collected via web scraping from the Google Play Store and automatically labelled based on user ratings, with ratings of 1-2 classified as negative sentiment and ratings of 4-5 as positive sentiment, whilst a rating of 3 was excluded as it was considered ambiguous. Following a preprocessing stage comprising cleaning, case folding, tokenisation, stopword removal, and stemming, as well as feature extraction using TF-IDF, 3,652 reviews were obtained with a training-to-test data split ratio of 80:20. The results indicate that positive sentiment dominates the dataset at 56.49%, whilst negative sentiment accounts for 43.51%. Analysis by application revealed that Kredivo was dominated by positive sentiment (68.20%), whilst Akulaku was dominated by negative sentiment (51.70%). The Naïve Bayes multinomial model achieved an accuracy of 84.13%, with average precision, recall, and F1-score values of 0.84, demonstrating good and balanced classification performance across both sentiment classes.
Abstract:This study discusses the implementation of the Naïve Bayes method to predict catering sales at PT.Negara Rasa Indonesia. The background of this study is based on the problem of suboptimal sales due to the absence of a structured…
tructured sales prediction system. The Naïve Bayes method was chosen because of its simplicity, speed, and ability to classify data with a high degree of accuracy. The data used in this study is historical sales data from the last two years, which has undergone cleaning, labeling, and transformation into four sales categories, namely very popular, popular, fairly popular, and less popular. The testing process was carried out using RapidMiner software by dividing the dataset into training data and test data at various ratios of 80:20. The test results showed a very high level of accuracy, with the highest value reaching 91.41%. These findings prove that the Naïve Bayes method is reliable for predicting catering sales, thereby assisting decision-making in more efficient sales management and planning at PT. Negara Rasa Indonesia.
Abstract:Penelitian ini mengkaji reaksi publik terhadap keputusan Mahkamah Konstitusi (MK) Indonesia yang mempertahankan batasan umur minimal 35 tahun untuk calon presiden dan wakil presiden. Dengan menggunakan metode Naïve Bayes…
s untuk menganalisis sentimen dari data Twitter, penelitian ini bertujuan untuk mengungkap persepsi publik terhadap regulasi ini. Analisis menunjukkan mayoritas sentimen negatif (90.9%), dengan hanya 6.6% sentimen positif dan 2.5% sentimen netral, menandakan ketidakpuasan yang dominan di kalangan publik. Akurasi analisis sentimen yang dihasilkan mencapai 67.98%, menegaskan efektivitas Naïve Bayes dalam konteks ini. Penelitian menghasilkann betapa pentingnya akan pembahasan lebih mendalam mengenai syarat pencalonan yang dapat mencerminkan aspirasi masyarakat agar mempertimbangkan aspek pengalaman dan kedewasaan. Dalam konteks yang lebih luas, temuan ini memberikan wawasan berharga tentang dinamika opini publik dan potensi revisi peraturan terkait, merekomendasikan kajian lebih lanjut untuk memahami dampak kebijakan tersebut terhadap struktur demokrasi Indonesia
Abstract:Dalam data mining, penelitian mengenai klasifikasi kelulusan mahasiswa sudah pernah dilakukan oleh peneliti lain. Sebagian besar penelitian tersebut difokuskan pada identifikasi variabel prediktor. Ada banyak penelitian…
dalam literatur terdahulu yang menjelaskan faktor-faktor apa saja yang dapat mensukseskan proses pengklasifikasian kelulusan mahasiswa. Pemilihan penggunaan algoritma Naive Bayes, dan pada penelitian ini didasarkan pada beberapa alasan, yaitu: Selain ketiga algoritma tersebut sama-sama mudah diimplementasikan dan sama-sama dapat memberikan hasil yang baik dalam kasus klasifikasi, ketiga algoritma tersebut juga mempunyai beberapa keunggulan masing-masing. Implementasi data training sebanyak 51 data dengan algoritma Naive Bayes berhasil memprediksi besarnya kelulusan mahasiswa dengan persentase keakuratan sebesar 74,67%, Sebanyak 184 mahasiswa sebagai data uji yang dihasilkan penelitian ini bahwa mahasiswa yang akan lulus tepat waktu sebanyak 42 mahasiswa atau sekitar 22,8% dari jumlah data testing dengan keakuratan sebesar 74,67%, Bagi Prang Studi Teknik Informatika berdasarkan hasil penelitian ini diharapkan dapat dimanfaatkan oleh pihak jurusan sebagai bahan pertimbangan bahwa dari jumlah data testing sebanyak 184 mahasiswa yang diperdiksi lulus tepat waktu sekitar 22,8% dan jumlah tidak tepat waktu 77,2%.
Abstract:This research aims to apply the Naïve Bayes method to determine the eligibility of receiving the Smart Indonesia Program (PIP) at the 017107 Kisaran Naga State Elementary School by analyzing 207 student data. The CRISP-DM…
DM approach was used through six stages: business understanding, data understanding, data preparation, modeling, evaluation, and implementation. The variables analyzed included means of transportation, KPS and KIP recipients, worth a pip, reasons for eligibility, number of siblings, distance from home to school, and parents' income.
The results showed that this method achieved 89% accuracy, 85% precision for the positive class, and 92% for the negative class. A total of 125 students (59.9%) were declared eligible to receive assistance, while 82 students (40.1%) did not meet the criteria. The Naïve Bayes method is effective in supporting decision-making for the provision of targeted educational assistance
Abstract:Education plays an important role in improving the quality of human life, but economic constraints often prevent many students from continuing their education. The Smart Indonesia Program (PIP) was launched to address these…
ese issues by providing educational assistance. However, in its implementation, the selection of PIP recipients at IT Al-Ikhsan Private Junior High School is still inaccurate. This research aims to classify the eligibility of PIP recipients using the Naive Bayes method. This method is applied to student data from the school's Dapodik in 2024 which consists of 265 students. The data is processed through CRISP-DM data mining stages, namely Business Understanding, Data Understanding, Data Preparation, Modeling, Evaluation, and Deployment. As a result, the Naive Bayes model showed an accuracy of 92.31% with a precision value for the “Yes” class of 89%, recall 100%, and F1-score 94%. In conclusion, variables such as means of transportation, KPS and KIP recipients, parents' income, and distance from home to school affect the eligibility of PIP recipients.
Abstract:The development of information technology provides solutions for increasing efficiency and accuracy in decision-making, such as in determining students eligible for BLT at SD Swasta IT ABI Husni. This study aims to implement…
ment the Naive Bayes algorithm to support a more objective BLT recipient selection process. The method used is CRISP-DM, starting from understanding the problem, data preparation, to model implementation. The data analyzed included type of residence, KPS recipients, parents' income, KIP recipients, number of siblings, distance from home and reasons for eligibility for BLT used were data from students of SD Swasta IT ABI Husni in the odd semester of 2024/2025, with a total of 137 data. The results of the study showed that the Naive Bayes algorithm was able to achieve an accuracy level of 98% with precision and recall of up to 100%, proving the effectiveness of the model in minimizing classification errors. In conclusion, the use of the Naive Bayes algorithm can help make decisions that are more targeted, transparent, and fair in the distribution of BLT.
Abstract:This research is motivated by the large number of prospective students who simply choose a major when they want to enter a vocational school without considering their abilities. The Decision Tree or C45 method is used because…
cause it is able to make decision trees that are easy to describe, and has a level of efficiency in handling discrete and numeric attribute data. While the Naive Bayes method is used because it has a high accuracy of results. This research was conducted based on data from students of SMK Muhammadiyah 10 Kisaran which contained questions about feelings of wrong majors, interests, and determinants of other majors. Data is divided into 2 labels, namely free labels (y) and bound labels (x). Followed by dividing the dataset into training data and testing data with a ratio of 70:30 in both methods to get the level of accuracy. From the results given, it can be seen that the C45 algorithm has an accuracy of 85% and the Naive Bayes algorithm has an accuracy of 26%. This shows that the C45 algorithm is more effective in classifying the available datasets compared to the Naive Bayes.
Abstract:Graduation marks the completion of a certain level of schooling. This study aims to predict the graduation of students at SDN 016528 BP Mandoge based on their abilities. The goal of this research is to reduce the rate of…
student failure to graduate by making predictions based on examination scores collected by the institution. The method used in this study is Naive Bayes, a technique in Data Mining that utilizes probability and statistics to predict future outcomes based on previous data. This method was chosen due to its advantage in predicting graduation rates from concrete data, ensuring the results are reliable and applicable for future predictions. The dataset used in this study includes graduation data for SDN 016528 BP Mandoge students for the 2019/2020 academic year, comprising 171 students, with 120 students used for training data and 51 students for testing data, achieving a model accuracy of 98%.