<![CDATA[Marriage is an important aspect of social life that is influenced by cultural changes and public opinion, especially in the digital age. Public opinion on marriage is now widely disseminated through social media, both from traditional and modern perspectives. This study aims to classify public opinions on marriage using the Support Vector Machine (SVM) and Multi-Layer Perceptron (MLP) algorithms. The data used consists of 1,003 comments collected from social media. The study was conducted using two different approaches: stemming and data augmentation, which involves increasing the training data by modifying the original data to improve model performance. The results show that in the first approach, SVM achieved an accuracy of 77%, while MLP improved from 75% to 76% without stemming. In the second approach, data augmentation without stemming provided a significant improvement in accuracy, with SVM reaching 93% and MLP reaching 96%. Wordcloud visualization also highlights the importance of removing stopwords to reduce noise in the data. These findings indicate that data augmentation is an effective strategy for improving the performance of opinion classification models. This research contributes to the field of social sentiment analysis using machine learning approaches and is expected to serve as a reference for formulating policies aimed at improving marriage quality and family stability in Indonesia.]]>
Copyrights © 2026
