Evolutionary Deep Learning for Social Media Keyword Optimization

Abstract

This paper proposes an innovative method for intelligent and automated keyword selection tailored for Twitter data analysis, specifically addressing public health surveillance. It integrates deep learning with evolutionary computing to generate effective keyword sets for data collection. By optimizing against metrics like tweet retrieval power, this approach significantly enhances the relevance and reach of collected social media information. Empirical findings demonstrate that this combined learning strategy outperforms traditional human-driven keyword selection in real-world scenarios.