Enhancing Sustainable Development Through Sentiment Analysis of Public Digital Resources: A PSO-LSTM Approach

In recent years, the global paradigm of sustainable development has gained prominence, emphasizing the need to address present challenges while safeguarding future generations’ resources and opportunities. This paradigm integrates environmental, social, and economic dimensions, aligning with the Sustainable Development Goals (SDGs). Organizations and nations worldwide are increasingly adopting sustainable development strategies to ensure long-term economic growth, social equity, and environmental preservation. Simultaneously, public management has witnessed a transformation, leveraging data mining and artificial intelligence to enhance decision-making efficiency. This research explores the intersection of sustainable development and public management innovation, focusing on the intelligent analysis of large-scale data. Specifically, it introduces a novel sentiment classification methodology, combining BERT word vectors and PSO-LSTM optimization, for user-generated textual data within public digital repositories. By analyzing public sentiment, this research empowers public management platforms to make more informed decisions, fosters transparency, and contributes to the realization of sustainable development goals. This study lays the groundwork for future research in sustainable eco-service platforms, encompassing diverse data types and advanced technologies to enhance public management and sustainable development efforts.