历史文化街区承载着城市的文化与共同记忆，是构建城市意象的重要载体。历史文化街区的场所感知一直是场所研究的热点议题。随着大数据与人工智能技术的发展，结合社交媒体数据，运用文本挖掘等机器学习方法研究历史文化街区的场所感知已日趋成熟。然而，现有研究多聚焦使用者的单一视角，或探讨使用者群体内部的感知差异；对于设计者视角，仍主要依赖传统建筑学理论，缺乏基于大数据方法的系统挖掘，也缺乏使用者与设计者之间场所感知的系统对比。本研究基于拉普卜特提出的理论，即使用者和设计者具有不同的环境感知模式，运用大数据和文本挖掘方法，在统一框架下对比两类群体的场所感知，旨在拓展该领域在多主体感知比较方面的研究路径。本研究首先系统梳理了基于传统建筑学视角与基于文本挖掘方法的场所感知相关研究，并以北京市历史文化街区为研究对象，使用微博社交媒体数据与建筑设计网站数据，对使用者与设计者的场所感知进行对比分析。具体而言，在使用者视角下，通过分析2019年北京五环内的微博文本数据，采用BERT情感分析模型和高频词提取，来挖掘使用者对场所的感知；在设计者视角下，通过分析谷德网的设计项目文本，运用LDA主题模型和高频词提取，来挖掘设计者对场所的专业感知。本研究发现，使用者倾向于用联想的方式来理解场所，其场所感知更多依赖于个体经验、情感记忆和社会互动；而设计师则倾向于用知觉的方式来感知场所，从宏观尺度出发，通过分析建筑形式、结构构造、材料运用等要素，对场所进行系统性的评估。此结果和拉普卜特的理论具有一致性。在此基础上，本研究进一步分析了混合式场所营造案例，总结优化场所记忆表达的新策略，并以北京典型历史文化街区——南锣鼓巷为背景，结合对该街区使用者和设计者场所感知的挖掘结果，进行了设计实践。本研究通过文本挖掘方法，深入比较使用者和设计者对历史文化街区的场所感知差异，从大数据与人工智能视角对拉普卜特的经典理论进行了实证补充。同时，也为设计者更好地理解与弥合使用者与设计者之间的感知差距提供了理论支持与方法参考，有助于历史文化街区场所记忆的保护与延续。

Historical and cultural districts, which carry a city's culture and common memory, serve as important carriers for shaping the "Image of the city". The place perception of such districts has always been a central topic in place-related research. With the development of big data and artificial intelligence technologies, studies leveraging social media data and applying machine learning techniques such as text mining to analyze place perception in historic districts have become increasingly mature. However, most existing studies primarily focus on the perspective of users, or explore perceptual differences within the user groups. In contrast, research from the perspective of designers remains largely grounded in traditional architectural theories, lacking systematic exploration through big data approaches, as well as the systematic comparison between the perceptions of users and designers.Based on the theory proposed by Rapoport that users and designers perceive environments differently, this study integrate big data and text mining methods within a unified analytical framework, aiming at expanding the research scope in the field of multi-subject place perception comparison.The study first systematically reviews prior research on place perception from both traditional architectural and text mining perspectives. It then selects historic districts in Beijing as the case study area and compares the place perceptions of users and designers using microblogging social media data and architectural design website data. Specifically, user perception is extracted by analyzing the text data from Weibo within Beijing’s Fifth Ring Road in 2019, using a BERT-based sentiment analysis model and high-frequency word extraction. In contrast, designer perception is mined by analyzing the text of the projects on the architecture website “gooood,” using LDA topic modeling and high-frequency word analysis.The results indicate that users tend to perceive places by association, with perceptions relies more on personal experiences, emotional memories, and social interactions; whereas designers tend to perceive places by perception, and conduct systematic evaluations from a macro perspective by analyzing elements such as architectural form, structural composition, and material application. The results show consistency with Rapoport’s theory.Building upon this analysis, the study further analyzes hybrid placemaking cases, and summarizes new strategies for optimizing the expression of place memory. Taking Nanluoguxiang, a typical historical and cultural district in Beijing, as a design context, this study conducts a practical design experiment informed by the comparative perception analysis of users and designers.By employing text mining methods, this study deeply compares the differences between users' and designers' place perceptions of historical and cultural districts, and offers empirical support for Rapoport’s theoretical model from the perspective of big data and artificial intelligence. Moreover, it provides methodological references and theoretical guidance for designers to better understand and bridge the perceptual gap between users and designers, ultimately contributing to the preservation and continuation of place memory in historic districts.