基于机器学习的服装店照明环境指标对氛围感知影响的研究  

Research on the Influence of Lighting Environment Indicators in Clothing Stores Based on Machine Learning on Atmosphere Perception

作　　者：王首一 熊晔 冯华 马浩然 WANG Shouyi;XIONG Ye;FENG Hua;MA Haoran(School of Architecture and Art,Hebei University of Engineering,Handan 056038,China;School of Software,Henan University,Kaifeng 450046,China)

机构地区：[1]河北工程大学建筑与艺术学院,河北邯郸056038 [2]河南大学软件学院,河南开封450046

出　　处：《照明工程学报》2024年第4期125-134,共10页China Illuminating Engineering Journal

基　　金：河北省社会科学基金项目(HB22TY018)。

摘　　要：研究个性化服装店的照明环境与氛围感知的影响规律,并根据不同服装风格提出相应的照明环境优化建议。本文以四种服装风格下的665家服装店作为研究对象,通过主客观调研收集数据,运用方差法、CRITIC赋权等方法对数据进行处理与分析,利用XGBoost归回算法建立照明环境氛围感知预测模型,采用SHAP法解释照明环境指标对氛围感知的贡献程度、相关性及影响规律。可以通过舒适性、活泼性和超然性三个氛围感知维度来评价服装店的照明环境所营造出的氛围感。水平照度是影响舒适性的关键变量,并与之呈负相关;重点照明系数是影响活泼性的关键变量,并与之呈正相关;色温是影响超然性的关键变量,并与之呈正相关。Study the influence patterns of the lighting environment on atmosphere perception in personalized clothing stores,and propose corresponding lighting environment optimization suggestions based on different clothing styles.This study takes 665 clothing stores across four clothing styles as research subjects.Data were collected through both objective and subjective surveys,and were processed and analyzed using variance analysis,CRITIC weighting,and other methods.AN XGBoost regression algorithm was used to establish a lighting environment atmosphere perception prediction model.The SHAP method was employed to interpret the contribution,correlation,and influencing patterns of lighting environment indicators on atmosphere perception.The atmosphere created by the lighting environment in clothing stores can be evaluated through three atmosphere perception dimensions:comfort,liveliness,and transcendence.Horizontal illuminance is the key variable affecting comfort and is negatively correlated with it;the accent lighting coefficient is the key variable affecting liveliness and is positively correlated with it;and color temperature is the key variable affecting transcendence and is positively correlated with it.

关 键 词：服装店 照明环境 氛围感知 XGBoost SHAP 

分 类 号：TU113.642[建筑科学—建筑理论]