引用本文:[点击复制]
[点击复制]
【打印本页】 【在线阅读全文】【下载PDF全文】 查看/发表评论下载PDF阅读器关闭

←前一篇|后一篇→

过刊浏览    高级检索

本文已被:浏览 1887次   下载 16 本文二维码信息
码上扫一扫!
基于深度学习的城市公园游客情感倾向 分析——以天津水上公园为例
罗俊杰,雷泽鑫,胡一可,王苗*,曹磊
0
作者简介:
摘要:
游客情感倾向是反映城市公园受欢迎程度与优劣势 的重要指标,可缓解公园游客与公园设计师、管理者之间由于 信息不对等带来的矛盾,为研究城市公园问题提供了新视角。 基于深度学习算法的文本情感倾向分析技术为公园评论数据 分析提供了准确高效的新手段。以天津水上公园为例,从新浪 微博等5个互联网平台采集2016—2019年的评论数据,基于 EASYDL深度学习平台实现评论数据分类与情感极性判断, 从整体情感倾向、时空分布特征和分类情感倾向三方面对评论 数据进行分类解析。研究发现:水上公园游客整体情感倾向偏 积极,评论数量与情感均值在时间和空间分布上均有较大差 异;在分类情感倾向分析中,使用者对水体景观与植物景观的 积极情感度较高,对服务设施与主题活动的消极情感度较高。 基于深度学习的城市公园游客情感倾向分析方法弥补了传统定 性分析方法的不足,可为城市公园环境评价和优化提升提供依 据,并为相关研究提供新思路。
关键词:  风景园林  情感倾向分析  深度学习  城市公园  网络评论大数据  时空分布特征
DOI:
基金项目:
Analysis of Tourists' Sentiment Tendency in Urban Parks Based on Deep Learning: A Case Study of Tianjin Water Park
LUO Junjie,LEI Zexin,HU Yike,WANG Miao,CAO Lei
Abstract:
: Tourist sentiment tendency is an important indicator reflecting the advantages, disadvantages and popularity of urban parks. It can alleviate the contradiction between tourists, park designers and managers due to information differences, and provides a new perspective for the study of urban parks. The text sentiment analysis based on deep learning algorithm provides an accurate and efficient new method for park review data analysis. Taking Tianjin Water Park as an example, the review data from 2016 to 2019 were collected from five internet platforms such as Sina Weibo. Based on EASYDL deep learning platform, the review data were classified and analyzed from three aspects: overall sentiment tendency, spatiotemporal distribution characteristics, and classified sentiment tendency. The results show that: the overall sentiment tendency of the water park tourists is positive, and the number of comments and the average sentiment value are significantly different in time and space distribution. In the classification sentiment tendency analysis, the positive sentiment degree of users to water landscape and plant landscape is higher, while the negative sentiment degree of service facilities and theme activities is higher. The study suggests that the analysis method based on deep learning makes up for the shortcomings of traditional qualitative analysis methods, and can provide a basis for the environmental evaluation and optimization of urban parks, and provide a new perspective for related research. Keywords: landscape architecture; sentiment tendency analysis;
Key words:  landscape architecture  sentiment tendency analysis  deep learning  urban parks  big data in online reviews  spatiotemporal distribution characteristics

京公网安备 11010802028240号

用微信扫一扫

用微信扫一扫