Sentiment and Emotional Trajectory Analysis to Customer Feedback on Saudi Arabian Food Delivery Apps

The rapid growth of online food delivery platforms in Saudi Arabia has led to a surge in user-generated feedback, necessitating automated systems to extract insights for service improvement. Traditional sentiment analysis often falls short in capturing complex emotional elements influencing customer satisfaction. This study presents an emotion-aware feedback analysis system to convert customer reviews into actionable suggestions for restaurant owners and food delivery services. Using a transformer-based deep learning model, the system classifies reviews at the sentence level. Batch-based inference and lightweight text preprocessing ensure computational efficiency. Detected emotional patterns are combined with domain-specific service rules to generate recommendations focusing on key areas like food quality and delivery performance. Geographic visualization contextualizes major platforms like Marsool and Talabat. Experimental evaluation shows the system effectively detects emotional trends, providing deeper insights than conventional sentiment polarity. The findings highlight the usefulness of emotion-aware NLP in practical applications, enabling data-driven decision-making for service enhancement in Saudi Arabia's food delivery market