Sentiment Urgency Emotion Detection for Business Intelligence

In recent years, the impact of social media on individuals’ lives has significantly grown. People use it for various purposes, such as finding inspiration, fostering discussions, and gaining information. Meanwhile, businesses leverage social media platforms to advertise their products and services, inform customers about future promotions, and maintain a direct connection with their target audience, regardless of location. Additionally, social media serves as a valuable source of insights into people’s well-being and security-related sentiments. This project introduces a learning model named Sentiment Urgency Emotion Detection (SUED) designed for mining social media data and analyzing opinions. SUED incorporates three classifiers, including sentiment analysis, urgency detection, and emotion classification. The goal is to train the model to enhance its accuracy and F1 score, thereby increasing the precision and the percentage of correctly predicted text. To demonstrate the model’s capabilities, it will be applied to a Twitter account associated with one of the largest supermarket chains in the UK, focusing on the detection of sentiments, emotions, and the level of urgency in the content.