基于文本挖掘技术识别土地管理领域未来方向的研究
Research on Identifying the Future Direction of Land Management Based on Text Mining Technology

立足生态文明时代和新型城镇化建设背景，围绕国家经济社会高质量发展和治理体系现代化要求，企业、研究机构和政府部门积极寻求新兴趋势和问题的方法，这些趋势和问题可能会影响其未来的运营环境。本文利用文本挖掘技术来调查土地管理部门的未来信号。通过系统回顾有关文本挖掘来检测未来信号的文献后，本研究建议使用隐含狄利克雷分布模型来增强对未来信号的解释。这项研究的发现突出了与土地利益及其记录相关的广泛问题，确定了17个未来信号主题，从缓解气候变化和使用卫星图像进行数据收集到标准化和参与式土地整理。研究表明，在使用自动化过程时，区分弱信号与潜伏信号，较强信号和强信号是具有挑战性的。本研究总结了土地管理领域的当前论述，并指出了当前哪些主题正在蓬勃发展。
Based on the background of the era of ecological civilization and new urbanization construction, and around the requirements of high-quality national economic and social development and modernization of the governance system, enterprises, research institutions and government de-partments are actively seeking ways to address emerging trends and issues that may affect their future operating environment. This paper uses text mining technology to investigate the future signal of Land Management Department. After a systematic review of the literature on text mining to detect future signals, this study suggests using the Latent Dirichlet allocation model to enhance the interpretation of future signals. The findings of the study highlighted a wide range of issues related to land interests and their records, and identified 17 future signal themes, ranging from climate change mitigation and the use of satellite imagery for data collection to standardization and participatory land consolidation. The research shows that it is challenging to distinguish weak signal from latent signal, strong signal from strong signal in the process of automation. This study summarizes the current discussion in the field of land management, and points out which topics are currently developing vigorously.