中国作为全球超特大城市最多的国家，其城市规划和治理近年来又面临来自人口老龄化、“双碳”目标和产业结构深度调整等宏观因素变化所带来的巨大挑战，而进一步推动绿色可持续的新型智慧城市建设依然是重要的应对方略，也是我国“十四五”期间创新驱动国家高质量发展战略实施的重要举措。针对全球范围内智慧城市技术热点及其演进模式的分析，对于我国当前智慧城市建设相关的战略规划和政策法规制定具有较为重要的决策参考价值；但是，目前国内外相关的研究均较少，尤其是针对智慧城市技术热点演进模式的分析基本处于空白。本文基于一个整合了传统主题建模、专利文本分析和主题强度演进的量化分析框架，尝试刻画出全球范围内智慧城市技术热点演进的知识图谱，这对于我国新时期促进智慧城市建设相关的政策制定或许具有一定的决策参考价值。智慧城市是一个典型的多学科交叉主题，本文尽管参考了多篇主流文献有关智慧城市的定义，专利检索也不可能完全覆盖所有智慧城市相关的技术领域；但是，本文针对一个交叉主题提出的专利检索策略和技术热点分析方法或许提供了一个科技情报挖掘的新思路。

As the country with the largest number of super-megacities in the world, China's urban planning and governance in recent years are facing great challenges brought about by the changes of macro factors such as population aging, "dual carbon" goal and deep adjustment of industrial structure etc. The construction of new smart cities is still one of the important approaches of the innovation-driven national high-quality development strategy during the “14th Five-Year Plan” period. The analysis of the global smart city technology hotspots and their evolution patterns has important reference value for the strategic planning and policy and regulation formulation of smart city in China. However, at present, there are few relevant studies at home and abroad, especially the analysis on the evolution pattern of smart city technology hotspot is almost blank. This paper puts forward an integrated framework that includes traditional topic modeling, patent text analysis and the evolution of topic-strength, and then more accurately portrays the knowledge map of the technology evolution on smart-city. This work reported by this paper may have a certain implication for the corresponding policy-making and strategy implementation in the new era. Smart city is a typical interdisciplinary topic. Although this paper refers to the definition of smart city in many mainstream literatures, patent search cannot completely cover all the technology fields related to smart city. However, the patent search strategy and technology hot spot analysis method proposed in this paper for a cross-topic may provide a new idea for intelligence gathering and analysis.

国家自然科学基金面上项目(72074081);广东省软科学研究计划重点项目(2019B101001002).