<![CDATA[The digital era has fundamentally changed news consumption patterns, with social media becoming the primary platform for information access used by 54% of Americans and 44% of Indonesians. This transformation creates significant challenges for researchers and communication practitioners in understanding public sentiment towards news in the digital era. This study aims to analyze public opinion sentiment towards news trends using the Random Forest algorithm. This study uses an experimental quantitative approach to analyze public opinion sentiment towards political news trends using the Random Forest algorithm. The study population consists of Twitter posts discussing political news topics from 2020 to 2025, with a sample of 1000 tweets selected through purposive sampling. The research instruments include hardware (AMD Ryzen 3 processor, 4GB RAM) and software (Python, Google Collaboratory), with data analysis techniques involving comprehensive preprocessing (cleaning, case folding, tokenization, stemming, stopword removal), TF-IDF transformation, and Random Forest implementation. The results show that the Random Forest model achieved an optimal accuracy of 81% at a 90:10 data split, with an accuracy range of 77-81% across various data split scenarios. This study concludes that Random Forest can be effectively implemented as a public sentiment monitoring instrument for governments, media, and public organizations, although larger datasets and multiple social media platforms are needed for better generalization..]]>
