基于NLP定制模型的游客感知研究——以重庆市鹅岭公园为例  

A Study on Tourists’Perception of Eling Park in Chongqing City Based on NLP Customized Model

作　　者：张汇雯 邓宏[1] 冯琬清 ZHANG Huiwen;DENG Hong;FENG Wanqing(School of Architecture and Urban Planning,Chongqing University,Chongqing,China,400044)

机构地区：[1]重庆大学建筑城规学院,重庆400044

出　　处：《园林》2023年第12期122-130,共9页Landscape Architecture Academic Journal

基　　金：国家自然科学基金项目“巴渝传统场镇景观价值识别及活态保护研究”(编号:51978094)。

摘　　要：游客感知对于了解人民需求、提升城市建设质量有着重要意义。以公园网络文本为数据训练NLP定制模型,更适用于风景园林领域的需求,使公园治理与设计更加智能与高效。基于深度学习平台,训练三个多标签文本分类、情感倾向分析、评论观点提取三个模型处理公园网络文本,从“时间—评价对象—评价对象下的感知要素”多层次分析公园游客情感倾向特征,挖掘重点感知要素。研究表明:(1)鹅岭公园游客感知整体积极性较高,6类评价对象中园外景观受关注度最高,自然景观与游客积极情绪成正比,设施配套消极情绪最高,停车位是亟需解决的问题。(2)在60个高频感知要素中,7个要素与游客积极情绪概率成显著正比,其中5个正相关,2个负相关。(3)采取“文本分类—高频词提取—情感分析”的分析顺序,可以挖掘词频低但有重要影响的感知要素。(4)NLP定制模型提供的属性级情感分析可以减少情感分析误差,使研究更准确。研究鹅岭公园游客情感与公园重点感知要素,为鹅岭公园的建设提升提出优化建议,为自然语言处理在风景园林中的应用提供了参考。The perception of tourists holds signifi cant importance in understanding people’s needs and enhancing the quality of urban development.Training an NLP customized model using park network text as data is more suitable for the demands of the land-scape architecture fi eld,enabling intelligent and effi cient governance and design of parks.Using a deep learning platform,three models were trained for multi-label text classifi cation,sentiment analysis,and comment viewpoint extraction to process park network text.A multi-level analysis of“time-evaluated object-perceived elements under the evaluated object”was conduct-ed to identify key perception features and explore the signifi cant elements.The research revealed the following fi ndings:(1)Overall,visitors’perception of Eling Park was predominantly positive,with the highest attention given to external landscapes among the six evaluated objects.Natural landscapes had the most positive impact on visitors’emotions,while negative emo-tions were associated with facilities and the need for parking space.(2)Among the 60 high-frequency perception elements,7 elements are signifi cantly proportional to the probability of tourists’positive emotions,of which 5 are positively and 2 are negatively correlated.(3)The analysis sequence of“text classifi cation-high-frequency word extraction-sentiment analysis”allowed the identifi cation of low-frequency perceived elements that had essential impacts on corresponding landscapes.(4)The attribute-level sentiment analysis provided by the NLP customized model reduced sentiment analysis errors,leading to more accurate research outcomes.This study examined the emotional perception of visitors and the key perceived elements of Eling Park,providing optimization suggestions for its development and contributing to the application of natural language processing in landscape architecture.

关 键 词：游客感知 网络文本分析 深度学习 情感倾向分析 评论观点抽取 

分 类 号：TU986[建筑科学—城市规划与设计]