Dynamic monitoring of public opinion on fertility intentions: based on the intersection of empirical and social media perspectives.

Introduction: In the context of China's rapid economic development and profound social changes, adjusting fertility policies has become a key national strategy. However, the long-term effectiveness of fertility incentives such as the three-child policy remains uncertain due to the lack of comprehensive economic and social support measures. Methods: This study innovatively utilizes Large Language Models (LLMs) to assist social media analysis, including intelligent noise removal and weighted sentiment analysis. By combining empirical data and a social media perspective, we developed the Dynamic Monitoring of Public Opinion on Fertility Intentions (DMPOFI). The aim of this paper is to analyze the factors influencing fertility intentions among Chinese women, providing theoretical support for policy formulation. The study analyzes 45,810 Weibo posts, all in Chinese, focusing on the public's views and sentiment toward fertility intentions. Through methods such as weighted sentiment analysis, LDA topic modeling, and TF-IDF, we gain insights into the changing public attitudes in recent years. Results: From the perspective of fertility intentions, the period of 2015-2016 was the most positive, while from 2017 to 2019, public attitudes toward fertility gradually became more rational and neutral. From 2020 to 2023, fertility intentions significantly declined, with neutral and negative emotions gradually dominating. The core factors influencing fertility intentions identified in this study include policy changes, social perceptions, economic pressures, work-life balance, cultural shifts, fluctuations in birth rates, and discussions on family planning. Discussion: Based on these findings, we propose a series of policy recommendations, urging the government to provide more flexible maternity leave policies, workplace support, and childcare services.