Valence and Arousal-Infused Bi-Directional LSTM for Sentiment Analysis of Government Social Media Management

Private entrepreneurs and government organizations widely adopt Facebook fan pages as an online social platform to communicate with the public. Posting on attract people’s comments shares is effective way increase public engagement. Moreover, comment functions allow users who have read posts express their thoughts. Hence, it also enables us understand users’ emotional feelings regarding that post by analyzing comments. The goal of this study investigate image exploring content pages. In order efficiently analyze enormous amount opinion data generated from media, we propose a Bi-directional Long Short-Term Memory (BiLSTM) can model detailed sentiment information hidden in those words. It first forecasts terms Valence Arousal (VA) values smallest unit text, later fuses into deep learning further whole text. Experiments show our achieve state-of-the-art performance predicting VA Additionally, combining BiLSTM results boost for media text analysis. Our method assist governments or other improve effectiveness operations through understanding opinions related issues.