Optimizing Sentiment Analysis on Twitter for Improved Customer Insights: Integrating Bagged CNN and Flamingo Search

Twitter helps you understand public opinion by revealing what people think, feel, and believe. This study provides a framework for businesses, particularly food delivery companies, to analyse social media data competitively and gain actionable insights. A novel combination of Flamingo Search Optimisation (FSA) and a one-dimensional bagged CNN is applied to Twitter data to improve customer satisfaction research. Our method uses advanced machine learning to analyse and interpret consumer attitudes and trends using Twitter's massive, real-time customer feedback stream. We found that sentiment analysis can now better predict consumer preferences and satisfaction. The study suggests developing more advanced optimisation algorithms, cross-platform sentiment analysis, real-time processing, and natural language processing techniques. Average kappa score: 0.9941, sensitivity: 0.9969, specificity: 0.9972, accuracy: 0.9971, precision: 0.9973. This study prepares companies to improve customer engagement and respond quickly to digital consumer sentiments.